Download (29Mb) - Glasgow Theses Service - University of Glasgow

Glasgow Theses Service http://theses.gla.ac.uk/

[email protected]

Podmore, Michelle (2000) Microbial targets of the humoral immune response in periodontal disease. PhD thesis http://theses.gla.ac.uk/3915/ Copyright and moral rights for this thesis are retained by the author A copy can be downloaded for personal non-commercial research or study, without prior permission or charge This thesis cannot be reproduced or quoted extensively from without first obtaining permission in writing from the Author The content must not be changed in any way or sold commercially in any format or medium without the formal permission of the Author When referring to this work, full bibliographic details including the author, title, awarding institution and date of the thesis must be given

http://theses.gla.ac.uk/

http://theses.gla.ac.uk/3915/

MICROBIAL TARGETS OF THE HUMORAL IMMUNE RESPONSE IN PERIODONTAL DISEASE

Michelle Podmore BSc

Thesis submitted for the degree of PhD to the Faculty of Medicine, University of Glasgow

Periodontology and Immunology Research Group, Glasgow Dental Hospital, University of Glasgow

ýýM. Podmore, October 2000

TEXT BOUND INTO

THE SPINE

Table of Contents

List of Tables IX

List of Figures XI

List of Abbreviations XVI

Acknowledgements XX

Declaration XXII

Summary XXIII

Chapter 1 General Introduction 1

1.1 Biological Structure of the Periodontium 1

1.1.1 Introduction 1

1.1.2 The gingiva 1

1.1.3 The oral epithelium 3

1.1.4 The sulcular epithelium 3

1.1.5 The junctional epithelium 3

1.1.6 Clinical features of the healthy gingiva 5

1.1.7 Clinical features of the diseased periodontium 5

1.2 The Classifications of Periodontal Disease 6

1.2.1 Normal gingiva 6

1.2.2 Gingival diseases 6

1.2.2.1 Plaque-induced gingivitis associated with local

factors 7

I

1.2.2.2 Plaque-induced gingivitis associated with local

factors and modified by specific systemic factors 8

1.2.2.3 Gingival diseases associated with systemic

diseases 10

1.2.3 Periodontitis 10

1.2.3.1 Chronic periodontitis 11

1.2.3.2 Aggressive periodontitis 12

1.3 The Microbiology of Periodontal disease 13


1.3.2 Porphyrornonas gingivalis 20

1.3.3 Actinobacillus actinomycetemcomitans 21

1.3.4 Prevotella intermedia 23

1.3.5 Bacteroidesforsythus 23

1.3.6 Spirochaetes 24

1.3.7 Microbial composition associated with health 25

1.3.8 Microbial composition associated with gingivitis 26

1.3.9 Microbial composition associated with

periodontitis 27

1.4 The Host Immune Response 27


1.4.2 The innate immune system 27

1.4.3 The inflammatory response 29

1.4.4 Complement 29

1.4.5 Prostaglandins 31

1.4.6 Cytokines in inflammation 31

1.4.6.1 Interleukin-1 32



1.4.6.4 Tumour necrosis factor-a 34

1.4.7 Adhesion molecules 35

1.4.7.1 The selectins 36

1.4.7.2 The integrins 37

II

1.4.7.3 The immunoglobulin superfamily 38

1.4.8 Polymorphonuclear leukocytes 42

1.4.9 The adaptive immune system 46

1.4.10 The humoral immune response 47

1.4.10.1 IgM 54

1.4.10.2 IgA 55

1.4.10.3 IgG 56

1.4.10.4 IgD 57

1.4.10.5 IgE 57

1.4.11 The role of the adaptive humoral immune

response in periodontal disease 58

1.4.12 The local antibody response 58

1.4.13 The systemic antibody response 61

1.4.14 The cell mediated immune response 64

1.4.15 T cell functions 67

1.4.1 1.4.15.1 The role of CD4+ T cells in periodontal disease 67

1.4.15.2 Gamma delta T cells 68

1.4.16 Natural killer cells 70

1.5 Cytokines in Periodontal Disease 71

1.5.1 Interleukin-1 71

1.5.2 Tumour necrosis factor 72

1.5.3 Interleukin-10 73

1.6 Important Antigens in Periodontal Disease 76


1.6.2 Outer-membrane proteins 83

1.6.3 Fimbriae 86

1.6.4 Lipopolysaccharide 89

1.6.5 Heat shock proteins 92

1.6.6 Leukotoxin 95

1.6.7 Cysteine proteases 97

1.6.8 Superantigens 99

III

1.6.9 Other strategies for identifying immunodominant

antigens 101

1.7 Pathogenesis of Periodontal Disease 103


1.7.2 Gingivitis 104

1.7.3 Periodontitis 105

1.8 Risk Factors 110

1.8.1 Introduction to risk factors 110

1.8.2 Age 110

1.8.3 Low socio-economic status 110

1.8.4 Smoking 111

1.8.5 Systemic disease 111

1.8.6 Other possible risk factors 112

1.8.7 Genetics 113

1.9 Aims and Objectives of Thesis 114


1.9.2 Aims of chapter 3: A comparison of the changes in the

humoral immune response to antigens of periodont al

pathogens following periodontal therapy 114

1.9.3 Aims of chapter 4: Immortalised B cells from

periodontal disease patients 115

1.9.4 Aims of chapter 5: Analysis of the target of

antibody secreting cells from diseased tissue of chronic

periodontitis patients by ELISPOT 115

1.9.5 Aims of chapter 6: Detection of antibody-bearing

plasma cells in periodontitis granulation and gingival

biopsy tissue 115

1.9.6 Layout of the thesis 116

Chapter 2 Materials and Buffers 117

2.1 Introduction 117

2.2 Materials 117

IV

2.2.1 A comparison of the changes in the

humoral immune response to antigens of

periodontal pathogens following periodontal

therapy: materials 117

2.2.2 Immortalised B cells from periodontal disease

patients: materials 120

2.2.3 Analysis of the target of antibody secreting cells from

diseased tissue of chronic peri. odontitis patients by

ELISPOT: materials 122

2.2.4 Detection of antibody-bearing plasma cells in

periodontitis granulation and gingival biopsy

tissue: materials 123

2.3 General Buffers 124

Chapter 3 A Comparison of the Changes in the

Humoral Immune Response to Antigens

of Periodontal Pathogens Following

Periodontal Therapy 129


3.2 Methods Part 1: Enzyme Linked

Immunosorbent Assay 136

3.2.1 Subjects and blood samples 136

3.2.2 Antigen preparation 137

3.2.3 Analysis of sera samples by Enzyme Linked


3.2.4 Statistical methods 140

3.3 Methods Part 2: Adsorptions 141

3.3.1 Subjects and blood samples 141


3.3.3 Adsorption of samples 142

3.3.4 Analysis of sera samples by Enzyme Linked


3.4 Results 143

V

3.4.1 Results of part I of the study: Enzyme Linked


3.4.1.1 Relationship between treatment and

antibody titres 164

3.4.1.2 Relationship of pocket depths and

antibody titres 165

3.4.1.3 Other parameters 165

3.4.2 Results of part 2 of the study: Adsorptions 165

3.5 Discussion 171

Chapter 4 Immortalised B Cells From Periodontal

Disease Patients 180


4.1.2 Epstein-Barr transformation 183

4.1.3 The CD40 system 184

4.2 Methods 186

4.2.1 Subjects 186

4.2.2 Cell culture 186


4.2.4 Lymphocyte preparation 187

4.2.5 Preparation of AET-treated sheep red

blood cells 188

4.2.6 T cell depletion 188

4.2.7 Expansion of the B cell numbers by the

Banchereau method 189

4.2.8 Enzymatic dissociation of granulation tissue 189

4.2.9 PEG fusion 190

4.2.10 Growth and preparation of bacteria 191

4.2.11 Analysis of hybridoma supernatants by the

Enzyme Linked Immunosorbent Assay 192

4.3 Results 192

4.3.1 L-cell culture 192

4.3.2 CRL-1668 cell culture 192

vi


4.3.4 Cell fusion 196

4.4 Discussion 196

Chapter 5 Analysis of the Target of Antibody

Secreting Cells From Diseased Tissue

of Chronic Periodontitis Patients By

ELISPOT 201


5.2 Methods 205

5.2.1 Patients and tissue samples 205


5.2.3 Enzymatic dissociation of granulation tissue 206

5.2.4 Enumeration of specific immunoglobulin

producing cells by ELISPOT 206

5.2.5 Statistical methods 207

5.3 Results 208

5.4 Discussion 218

Chapter 6 Detection of Antibody-Bearing Plasma

Cells in Periodontitis Granulation and

Gingival Biopsy Tissue 223


6.2 Methods 228

6.2.1 Preparation of Fab fragments of goat

anti-human IgG 228

6.2.2 Subjects and samples 229

6.2.3 Method 1 229

6.2.3.1 Bacteria preparation 229

6.2.3.2 Immunohistochemistry 230

6.2.4 Method 2 231

6.2.4.1 Antigen preparation 231


VII

6.2.5 Method 3 232

6.2.5.1 Antigen preparation 232

6.2.5.2 Biotinylation of bacteria and

bacterial sonicates 233

6.2.5.3 Biotinylation of purified antigens 234


6.2.6 Data collection and analysis 235

6.3 Results 236

6.4 Discussion 260

Chapter 7 Methodological Considerations

and General Conclusions 271

7.1 Methodological considerations 271

7.2 Conclusions 271

7.3 Suggestions for future research 276

7.3.1 Examination of the effect of treatment on the

humoral immune response 276

7.3.2 Further work on monoclonal antibody production 276

7.3.3 Detection of antibody-bearing plasma cells

in granulation and gingival tissue 276

7.4 Concluding remarks 277

Bibliography 278

List of publications 369

VIII

List of Tables

Table Title Page

Table 1.1 Major Adhesion Molecules 41

Table 1.2 The Role of Cytokines in Periodontal Disease 75

Table 3.1 Polyvalent immmunoglobulin titres before and after

treatment and the differences between these 144

Table 3.2 Polyvalent Immunoglobulin, IgG and IgA titres before and

after treatment and comparisons of these for P. gingivalis,

A. actinomycetemcomitans, P. interinedia, B. forsythus and

T. denticola 152

Table 3.3 The correlation between change in pocket depth and

change in antibody titre following treatment 155

Table 3.4 To show the correlation between pre-treatment pocket

depth and pre-treatment antibody titre 158

Table 3.5 To show the correlation between length of treatment and

change in antibody titre following treatment 161

Table 3.6 The median percentage reduction in antibody titre

following adsorption compared to sham adsorbed sera 166

Table 5.1 To show the median number of ELISPOTs enumerated

against the periodontal bacteria tested 215

Table 5.2 To show the median difference in the number of

ELISPOTs enumerated for the two different cell

concentrations 216

Table 6.1 Median figures for cells per field bearing antibody to the

whole bacterial cells and purified antigens tested.

Magnification x200 for granulation tissue sections taken

from patients with chronic periodontitis 247

IX

Table 6.2 Median figures for cells per field bearing antibody to the

whole bacterial cells and purified antigens tested.

Magnification x200 for gingival tissue sections taken from

patients with chronic periodontitis 247

Table 6.3 Median differences in the number of cells per field for the

micro-organisms and purified antigens tested in the

granulation and gingival tissue samples. Magnification

x200 for granulation and gingival tissue sections taken


Table 6.4 Mean figures for the percentage of the plasma cells per

field specific for the antigens tested. Magnification x200

for granulation and gingival tissue sections taken from


Table 6.5 Mean figures for the percentage of plasma cells per field

specific for the whole bacterial cells and antigens tested for

each of the patients individually. Magnification x200 for

granulation and gingival tissue sections taken from patients

with chronic periodontitis 254

X

List of Figures

Figure Title Page

figure 1.1 The structure of IgGI 48

figure 1.2 The pristine gingiva 107

figure 1.3 The healthy gingiva 107

figure 1.4 The early gingival lesion 108

figure 1.5 The established gingival lesion 108

figure 1.6 The advanced lesion 109

figure 3.1 Plot of polyvalent immunoglobulin titres pre- and post-

treatment against P. gingivalis 146


treatment against P. gingivalis OMP 146


treatment against the ß segment of RgpA 147

figure 3.4 Boxplot of the difference between pre- and post-treatment

antibody titres against LPS 150

figure 3.5 Plot of IgG titres pre- and post-treatment against P.

gingivalis 153

figure 3.6 Plot of IgA titres pre- and post-treatment against P.

gingivalis 153

figure 3.7 Plot of change in pocket depth against change in antibody

titre for B. forsythus 156

figure 3.8 Plot of change in pocket depth against change in antibody

titre for P. gingivalis 156

figure 3.9 Plot of pre-treatment pocket depth against pre-treatment

antibody titre for A. actinomycetemcomitans 159

figure 3.10 Plot of length of treatment against change in antibody

titre for heat shock protein 60 of P. gingivalis 162

figure 3.11 Plot of length of treatment against change in antibody

titre for P. gingivalis 162

XI

figure 3.12 Plot of the median percentage reduction in antibody titre

against P. gingivalis following adsorption against P.

gingivalis, A. actinomycetemcomitans, B. forsythus and P.

intermedia when compared to the titre observed against

the sham adsorbed sera 167


against A. actinonrycetemcomitans following adsorption

against P. gingivalis, A. actinomycetemcomitans, B.

forsythus and P. intermedia when compared to the titre

observed against the sham adsorbed sera 168


against B. forsythus following adsorption against P.





against P. intermedia following adsorption against P.




figure 4.1 L cells 194

figure 4.2 Heteromyeloma cells 194

figure 4.3 Proliferating cells at day 3 of culture 195

figure 4.4 Proliferating cells at day 11 of culture 195

figure 5.1a P. gingivalis ELISPOT 209

figure 5.1b Control 209

figure 5.2a A. actinomycetemcomitans ELISPOT 210


figure 5.3a P. intermedia ELISPOT 211


figure 5.4a B. forsythus ELISPOT 212


XII

figure 5.5 Repeated measures boxplot for the addition of 108 B

lymphocytes 213

figure 5.6 Repeated measures boxplot for the addition of 5x107 B

lymphocytes 213

figure 6.1 a Antibody-bearing cells specific for P. gingivalis in a

granulation tissue sample 237

figure 6.1b Control for P. gingivalis 237

figure 6.2a Antibody-bearing cells specific for A.

actinomycetemcomitans in a granulation tissue sample 237

figure 6.2b Control for A. actinomycetemcomitans 237

figure 6.3a Antibody-bearing cells specific for P. intermedia in a


figure 6.3b Control for P. intermedia 238

figure 6.4a Antibody-bearing cells specific for B. forsythus in a


figure 6.4b Control for B. forsythus 238

figure 6.5a Antibody-bearing cells specific for T. denticola in a


figure 6.5b Control for T. denticola 239

figure 6.6a Antibody-bearing cells specific for E. coli in a


figure 6.6b Control for E. coli 239

figure 6.7a Antibody-bearing cells specific for HSP60 (of P.

gingivalis) in a granulation tissue sample 240

figure 6.7b Control for HSP60 (of P. gingivalis) 240

figure 6.8a Antibody-bearing cells specific for leukotoxin (of A.

actinomycetemcomitans) in a granulation tissue sample 241

figure 6.8b Control for leukotoxin (of A. actinomycetemcomitans) 241

figure 6.9a Antibody-bearing cells specific for the ß segment of

RgpA (of P. gingivalis) in a granulation tissue sample 242

figure 6.9b Control for the ß segment of RgpA (of P. gingivalis) 242

XIII

figure 6.10 Plot of number of specific antibody-bearing cells per field

against different concentrations of whole fixed bacteria

used. Magnification x200 for granulation tissue sections


figure 6.11 Plot of number of antibody-bearing cells per field against

different concentrations of purified antigen.

Magnification x200 for granulation tissue sections from


figure 6.12a Pie chart to show the percentage of plasma cells specific

for each antigenic target tested in the granulation tissue

sample of patient 1 255

figure 6.12b Pie chart to show the percentage of plasma cells specific

for each antigenic target tested in the gingival tissue




















XIV



sample of patient 5



sample of patient 5

259

259

xv

List of Abbreviations

Aa Actinobacillus actinomycetemcomitans

A. actinonrycetemcomitans Actinobacillus actinomycetemcomitans

Aa OMP Actinobacillus actinomycetemcomitans

outer-membrane proteins

ADCC Antibody-dependent cell mediated

cytotoxicity

APC Antigen presenting cell

AT After periodontal treatment

ATCC American Type Culture Collection

A. viscosus Actinomyces viscosus

Bf Bacteroides forsythus

B. forsythus Bacteroidesforsythus

Bf OMP Bacteroidesforsythus outer-membrane

proteins

BOP Bleeding on probing

BT Before periodontal treatment

CB Coating buffer

CD Cluster of differentiation antigen(s)

CD3 Integral part of the T cell receptor

complex

CD4 Marker for T-helper cell(s)

CD8 Marker for cytotoxic T cell

CD45RA Marker for naive and resting memory T

cell

CD45RO Marker for activated memory T cell

C. I. Confidence interval

CTL Cytolytic T lymphocyte

C. rectus Campylobacter rectus

DNA Deoxyribonucleic acid

EBV Epstein Barr virus

XVI

Ec Escherichia coli

E. coli Escherichia coli

ELAM-1 Endothelial Leukocyte Adhesion

Molecule-1

ELISA Enzyme linked immunosorbent assay

ELISPOT Enzyme linked immunospot

E. saburreum Eubacterium saburreum

FAA Fastidious anaerobe agar

FCS Foetal calf serum

FDC Follicular dendritic cell

F. nucleatum Fusobacterium nucleatum

GC Germinal centre

GCF Gingival crevicular fluid

GlyCAM-I Glycosylated Cellular Adhesion

Molecule-I

HEL Hen egg lysozyme

H. influenzae Haemophilus influenzae

HIV Human immunodeficiency virus

HPT Hygiene phase therapy

HPRT Hypoxanthine phosphoribosyl transferase

HRP Horseradish peroxidase

HSP Heat shock protein

HVJ Hemagglutinating virus of Japan

ICAM (-1,2,3) Intercellular Adhesion Molecule

IgA Immunoglobulin A

IgD Immunoglobulin D

IgE Immunoglobulin E

IgG Immunoglobulin G

IgM Immunoglobulin M

IL Interleukin

INF-y Interferon gamma

kDa Kilodalton

XVII

LAD Leukocyte adhesion deficiency

LBP Lipopolysaccharide binding protein

LFA (-1,2,3) Lymphocyte function-associated antigen

LPS Lipopolysaccharide

LOS Lipooligosaccharide

LTX Leukotoxin

M. avium Mycobacterium avium

MAC Mycobacterium avium complex

MadCAM-1 Mucosal Addressin Cellular Adhesion

Molecule-1

MHC Major histocompatability complex

mRNA Messenger Ribonucleic acid

NADPH Nicotinamide adenine dinucleotide

phosphate (reduced form)

NK cell Natural killer cell

OMP Outer-membrane protein

PALs Periarteriolar lymphoid sheaths

P. D. Probing pocket depth

PECAM Platelet Endothelial Cell Adhesion

Molecule

Pg Porphyromonas gingivalis

P. gingivalis Porphyromonas gingivalis

Pg OMP Porphyromonas gingivalis outer-

membrane proteins

Pi Prevotella intermedia

P. intermedia Prevotella intermedia

Pi OMP Prevotella intermedia outer-membrane

proteins

PGE2 Prostaglandin E2

PMN Polymorphonuclear leukocyte

P. nigrescens Prevotella nigrescens

RNA Ribonucleic acid

XVIII

rRNA Ribosomal Ribonucleic acid

RTX Repeats in toxin

SRBC Sheep red blood cells TCR T cell receptor

Td Treponema denticola

T. denticola Treponema denticola

TH T-helper cell subset

TNF-a Tumour necrosis factor alpha

T. pectinovarum Treponema pectinovarum

T. socranskii Treponema socranskii

T. vincentii Treponema vincentii

V. alcalescens Veillonella alcalescens

VCAM-1 Vascular Cell Adhesion Molecule-1

VLA-4 Very Late Antigen-1

XIX

Acknowledgements

I would like to express my sincere thanks to the following people who have helped me

during my time at Glasgow Dental Hospital both in my research and the preparation of

this thesis:

My Supervisor Professor Denis Kinane for all of the help and encouragement that he

has given me during my 3 years at Glasgow Dental School. For editing my thesis and

giving me the opportunity to acquire many experiences and skills during my time in

his research group.

Dr. Ivan Darby and Graeme Marshall for the use of their clinical samples.

Dr. David Lappin for giving me so much of his valuable time, help and advice.

Miss Siobhan McHugh for all her advice on the statistics in this thesis and for

spending time with me to ensure that I had developed a good understanding of the

topic.

Mr Alan Lennon for culturing and maintaining the micro-organisms used in the

studies described in this thesis.

Mr Duncan Mackenzie and the technical and administrative staff on level 9.

Mr Richard Foy for his help and advice on photography.

Mr Jim Daly for all his help over the last 3 years.

Helen Marlborough for all her help with my literature searches and use of Reference

Manager.

Beverly Rankin and Christine Leitch in the James Ireland Library.

xx

Dr. Penelope Hodge, Graeme Marshall, Takehiko Kubota, John Mooney and

particularly my father Colin Podmore for proof reading and editing this thesis.

I would like to thank all my friends in the Dental Hospital that have given me support

and encouragement and kept me laughing during my write-up.

I would like to thank Scott for all his support and tolerance, you are a man with a lot

of patience.

Finally, I would like to thank my parents Pat and Colin without whom this would

never have been possible either financially or emotionally. Their constant support,

encouragement and words of wisdom have kept me going and it is therefore, to them

that I dedicate this thesis.

xx'

Declaration

This thesis is the original work of the author

I l1 ý/

ci' Michelle Podmore BSc

XXII

Summary

Periodontitis is a ubiquitous infectious disease severely affecting 10% of the

population. There are many areas of our knowledge regarding this disease that are

incomplete. In particular, the microbial, inflammatory and immunological aetiology

and pathogenesis of this disease remain the subject of many current investigations.

Substantial literature exists on the humoral immune response against micro-organisms

present within periodontal pockets and tissues. However, the literature is complex

and it is difficult to elucidate due to the large number of putative periodontal

pathogens, the large array of immunogenic antigens present on the surface of these

micro-organisms and the potential cross reactive nature of them. In addition, there is

enormous variation in both individual immune responses and the timing of a serum

sample collection with respect to treatment stage and disease course. Identification of

the immunodominant antigen or antigens involved in the disease process of

periodontitis would be extremely informative, as it would pin point the important

pathogens relevant to periodontal disease and identify the specific bacterial antigens

of importance. In addition, it may furnish us with new approaches to diagnosis, aid

the selection of potential vaccine candidates and suggest new therapeutic possibilities.

This thesis reports on investigations carried out to identify the important antigens

inducing a humoral immune response in periodontal disease. A number of different

techniques were used. The first study examined the systemic immune response,

specifically the effects of treatment on the immune response of patients with chronic

periodontitis against a panel of periodontal pathogens and a large panel of antigen

preparations from these bacteria. The results of the study indicated that there was no

difference in antibody titre following treatment against any of the putative periodontal

antigenic targets tested. These results conflict with many published reports, however,

differences in factors such as length of treatment and antigen preparation often make

the comparison of different studies futile.

XXIII

An investigation to examine the phenomenon of cross-reactivity between antibodies

induced against putative periodontal pathogens was the second part of this overall

project. The results suggested a high proportion of cross-reactivity between the 4

periodontal pathogens P. gingivalis, A. actinomycetemcomitans, P. intermedia and B.

forsythus. Results indicating such a high cross-reactivity between these micro-

organisms prove that the issue of cross-reactivity is an important one that should be

considered when carrying out any immunological and microbiological investigations

in this field.

Following the above investigations of the systemic immune response the remainder of

this thesis is a report of three studies examining the local immune response in

periodontal disease. The study reported in chapter 4 was an attempt to produce human

monoclonal antibodies from immortalised antibody-bearing cells. These were isolated

from the granulation tissue of diseased sites and the peripheral blood of patients with

chronic periodontal disease. The aim of this study was to produce monoclonal

antibodies against periodontal pathogens to be able to examine the antigenic target of

the antibodies produced in more detail and to assess which bacteria induced humoral

responses. Sequencing of the antigen binding domain of a monoclonal antibody

produced in this way could indicate which antigen on the surface of a micro-organism

is immunogenic and the target of the immune response in a particular disease. The

results were disappointing in terms of monoclonal antibody production, however,

antibodies specific for putative periodontal pathogens were initially produced which

were indicative that the technique has potential for this type of study.

An alternative approach was, therefore, undertaken to investigate the local immune

response in periodontal disease. Chapter 5 describes the study that was carried out to

examine the target of antibody-bearing cells isolated from granulation tissue removed from patients with chronic periodontal disease. This was investigated utilising

Enzyme Linked Immunospot (ELISPOT), a technique which detects antibody-

secreting cells directly. The results of the study indicate that there are antibody- bearing cells in diseased periodontal lesions specific to the pathogens tested, and there

is no significant difference in the number of antibody-bearing cells detected against

the 4 micro-organisms. XXIV

The final study set out to detect and enumerate antibody-bearing cells in tissue

samples and this was accomplished by the development of an innovative and unique

technique. The technique used was an immunohistochemical technique developed for

this project. The results of this experiment were in agreement with those of the

ELISPOT technique, that there was no significant difference in the number of cells

detected against the antigenic targets tested. When the number of cells detected

against the antigenic targets were compared in the granulation and gingival tissues

sections there were significantly more cells detected against P. gingivalis, A.

aetinomycetemcomitans and P. intermedia in the granulation tissue sections. When

the numbers of cells were expressed as percentages of the total number of infiltrating

plasma cells, however, this difference was not seen. To have the ability to express the

numbers of cells specific for a particular antigen as a percentage of the total infiltrate,

and in particular the total plasma cell infiltrate, could have major effects in the

determination of the importance of a particular micro-organism and antigenic target

and in the overall magnitude of a specific response.

The future identification of the target of the humoral immune response in periodontal

disease could have wide-ranging effects, not only in the progression of research and

knowledge but also in terms of diagnostics and therapeutics. Numerous reports have

suggested that elevations in humoral immune responses to micro-organisms are

associated with disease severity (Genco et al., 1980b; Mouton et al., 1981; Tolo and

Brandtzaeg, 1982; Listgarten et al., 1981; Ebersole et al., 1982b). Thus the discovery

of the target of the humoral immune response is of utmost importance in diagnosis of

disease activity and the possibilities of enhancing or modifying the humoral immune

response to achieve treatment outcome benefits.

xxv

Chapter 1. General Introduction

1.1 Biological Structure of the Periodontium

1.1.1 Introduction

A definition of the periodontium is "the tissues investing and supporting the teeth"

(Hassell, 1993). This includes the alveolar bone, root cementum, periodontal

ligament, and the gingivae. Together these form a functional unit (Lindhe and

Karring, 1997), which can be reversibly or irreversibly disrupted by the inflammatory

mechanisms of disease. Periodontal disease is a general term for the inflammatory

reactions affecting the periodontium. From the above description, it can be seen that

there are a number of different tissue types from a number of different development

origins making up the periodontium, thus disease affecting any one of these tissues is

strictly speaking a disease of the periodontium (Kinane and Davies, 1990). In the

context of this thesis the term periodontal disease will be restricted to denote

gingivitis and periodontitis, as classified in section 1.2.

1.1.2 The gingiva

The gingiva is the part of the oral mucosa that covers the alveolar processes of the

jaws and the cervical portions of the teeth. Anatomically, it is divided into three

distinct areas; the marginal, attached and interdental areas. The tissues collectively

termed the gingiva are said to belong to the oral mucous membrane and the

periodontium (Schroeder and Listgarten, 1997). Clinically, the gingiva is regarded as

a combination of epithelial and connective tissues. These make up the mucosa that is

around the teeth of the complete deciduous or permanent dentition and is attached to

both teeth and alveolar processes (Schroeder and Listgarten, 1997).

The gingiva is normally pink in colour, has a scalloped outline, a firm texture,

stippling, and is demarcated apically from the oral mucosa by the mucogingival line.

The colour of the gingivae may vary due to various amounts of melanin pigmentation.

The mucogingival line normally resides approximately 3-5mm apical to the level of

the alveolar crest. Attached keratinsed gingiva extends coronally from the

mucogingival line and is firmly bound to the underlying periosteum by collagen

fibres. The corono-apical width of the attached gingiva can vary significantly from

tooth to tooth. The free gingival margin, which is normally about 1.5mm in the

corono-apical dimension, surrounds but is not attached to each tooth. In health the

gingiva completely fills the embrasure space between the teeth, and this part is termed

the papilla (Wennström, 1988).

The gingival sulcus consists of the tooth surface on one side and sulcular epithelium

of the free gingiva on the other. It is a shallow crevice, and in fully developed teeth is

lined coronally with sulcular epithelium, the non keratinised extension of the oral

epithelium. The bottom of the sulcus is formed by the coronal surface of the

junctional epithelium.

The attached gingiva is continuous with the free gingiva and its apical extension leads

to the mucogingival junction. At this point it is continuous with the relatively loose

and movable alveolar mucosa. The attached gingiva is tightly attached to the

underlying alveolar bone and demonstrates comparative immobility relative to the

underlying tissue. The width of the attached gingiva differs among different teeth

(Bowers, 1963).

The interdental gingiva occupies the interproximal space beneath the area of tooth

contact. The interdental gingiva can be pyramidal in shape i. e. one papilla's tip is

immediately beneath the contact point, or have a `col' shape i. e. a valley-like

depression that connects a facial and a lingual papilla and conforms to the shape of the

interproximal contact (Cohen, 1959).

The marginal gingiva is comprised of three areas of epithelium; the oral epithelium

which faces the oral cavity, the sulcular epithelium which faces the tooth without any

contact with the tooth surface and the junctional epithelium which is in contact with

the tooth surface. The principal cell type of the gingival and oral epithelium is the

keratinocyte, which constitutes more than 90% of the gingival epithelium. Other cell

2

types present include Langerhans cells (DiFranco et at., 1985), Merkel cells (Ness

Morton and Dale, 1987) and melanocytes (Schroeder, 1969a). Keratinisation is a

process that involves a sequence of biochemical and morphologic events that occur in

the cell during its migration from the basal layers towards the surface (Schroeder,

1969b). The gingival epithelium is able to express three types of surface

differentiation; keratinisation, where the surface cells form scales of keratin and lose

their nuclei, parakeratinisation, where the cells of the superficial layers retain their

nuclei but there is no granular layer present, and non-keratinisation where the cells

preserve their nuclei but all signs of keratinisation are absent.

1.1.3 The oral epithelium

The crest and the outer surface of the marginal gingiva and the surface of the attached

gingiva are covered by the oral epithelium. This epithelium is made up of cells that

are stratified squamous type, and exhibit keratinisation or parakeratinisation or both.

The epithelium is bound to connective tissue by a basal lamina (Stem, 1965). In

health the border between the oral epithelium and the connective tissue is

characterised by deep epithelial ridges or rete pegs which are separated from each

other by connective tissue papillae.

1.1.4 The sulcular epithelium

The sulcular epithelium lines the gingival sulcus and is a thin, non-keratinised

stratified squamous epithelium displaying rete pegs.

1.1.5 The junctional epithelium

The junctional epithelium is also known as the dento-gingival epithelium. It is

characterised by a collar-like band of stratified squamous epithelium and can range in

length from 0.25 to 1.35mm. The junctional epithelium is attached to the tooth

surface by a basal lamina (Listgarten, 1966). Unlike the oral and sulcular epithelium,

the junctional epithelium lacks rete pegs (Lindhe and Karring, 1997).

3

The lamina propria consists of supra-alveolar fibre apparatus, blood and lymphatic

vessels and nerves (Schroeder and Listgarten, 1997). The fibre apparatus is a dense

network of fibre bundles which make up 55-60% of the connective tissue volume.

The fibres are densely populated by fibroblasts and consist of collagen type I and III

(Narayanan and Page, 1976). Mast cells are regularly found in this area and

lymphocytes, monocytes and macrophages can also be evident. The supra-alveolar

fibre apparatus plays an important role in attaching the gingiva to the teeth and bone,

and also provides a dense framework that gives the rigidity and biochemical resistance

to the gingiva (Schroeder and Listgarten, 1997). As well as providing this important

rigid framework, the fibre apparatus also protects the cellular defences that are located

at the dento-gingival interface (Schroeder, 1986).

The gingival lamina propria or connective tissue, is highly vascularised. Branches of

the main arteries supplying the blood to the maxilla and the mandible reach the

gingiva from 3 sites; the interdental septa, the periodontal ligament and the oral

mucosa (Castelli, 1963, Saunders and Röckert, 1967). The blood vessels of the

gingiva are arterioles, capillaries and small veins. The nutritional supply for the

gingival epithelium is provided by thin capillaries that terminate immediately below

the basement membrane. The lymphatic drainage of the periodontal tissues involves

the lymphatics of the connective tissue papillae. The lymph progresses into the

collecting network of the larger lymph vessels and is drained into the regional lymph

nodes of the head and the neck, before entering the blood circulation. Lymphatics just

below the junctional epithelium extend into the periodontal ligament and accompany

the blood vessels. High endothelial venules are not present when the gingiva is

healthy, however, even in the early stages of experimental gingivitis, high endothelial

venules are seen. It is considered their presence may be induced by mediators of

inflammation. The reason for their presence is not confirmed, however, it may

indicate a process of selective homing and migration of lymphocytes to the gingiva on

infection (Wynne et al., 1988).

4

1.1.6 Clinical features of the healthy gingiva

In `normal' appearance the gingiva is pink in colour, firm in consistency and the

gingival margin has a clear scalloped outline. It does not bleed when gently probed

and fills the entire space between adjacent teeth (Ainamo and Löe, 1966; Wennström,

1988). Characteristically in health, the depth of the gingival sulcus is minimal at the

microscopic level, with the alveolar bone located 1mm apical to the cementoenamel

junction (Eliasson et al., 1986). The oral surface is covered by oral epithelium that is

keratinised and fused with the junctional epithelium. The junctional epithelium is

weakly attached to the tooth surface by hemidesmosomes. The oral and junctional

epithelia are supported by the collagenous structures of the underlying connective

tissue. Directly below the junctional epithelium is a microvascular plexus containing

numerous venules (Egelberg, 1967).

1.1.7 Clinical features of the diseased periodontium

The discussion of the pathogenesis and histopathology of gingivitis and periodontitis

can be found in section 1.7 of this thesis. As discussed there, the inflammatory and

immune reactions that results from the hosts response to the plaque microflora and its

products are visible both histopathologically and clinically in the affected

periodontium.

Periodontal disease is broadly classified into two distinct entities; gingivitis and

periodontitis. Gingivitis refers to the pathological changes which are confined to the

gingivae. Clinically this disease entity is characterised by a change in colour from

pink to red of the gingivae, as well as a change in texture and appearance of the

gingivae as it swells. The gingivae becomes more prone to bleeding on gentle probing

and gingival sensitivity may be experienced, however, this disease is largely reversible

(Löe et al., 1965). Periodontitis, affects the deeper attachment apparatus including the

alveolar bone, periodontal ligament and the cementum, resulting in loss of periodontal

support, and is largely irreversible. It is frequently associated with the presence of

periodontal pockets, bleeding on probing, and bone loss is a pathognomonic feature.

5

1.2 The Classifications of Periodontal Disease

1.2.1 Normal gingiva

A normal healthy gingiva is characterised as described in 1.1.6. In histological terms

a normal gingivae is free from inflammation and accumulation of inflammatory cells

is not significant. As described by Kinane and Lindhe (1997), two different types of

`healthy' gingivae can be described. The `pristine' gingiva has all the characteristics

described above including little or no inflammatory infiltrate. This condition can only

be achieved in humans by several weeks of experimental conditions requiring

supervised and meticulous daily plaque control. The second type is the `clinically

healthy' gingiva, which again has the features as described above, but histologically

an inflammatory infiltrate is seen. This picture of health is clinically seen in everyday

situations.

1.2.2 Gingival diseases

The following classifications of periodontal diseases and conditions is based on the

recent classification system developed at the 1999 International Workshop for the

Classification of Periodontal Diseases and Conditions.

To define all the various clinical entities that effect the gingiva as `gingivitis' is

considered too restrictive and confusing. The term `gingival diseases' has, therefore,

been chosen as it provides a more comprehensive and encompassing definition of all

the different entities that effect the gingiva. The following characteristics are common

to all gingival diseases: (i) the clinical signs are confined to the gingiva; (ii) bacteria is

present and its role is in either initiation or exacerbation of the severity of the lesion;

(iii) there are clinical signs of inflammation and these include enlarged gingival

contours due to oedema or fibrosis (Mühlemann and Son, 1971; Poison and Goodson,

1985) colour transition from pink to a red and/or bluish-red hue (Mühlemann and Son,

1971; Polson and Goodson, 1985), elevated sulcular temperature (Haffajee et al.,

1992; Wolff et al., 1997), bleeding on probing (Mühlemann and Son, 1971; Löe et al.,

1965, Greenstein et al., 1981; Engelberger et al., 1983) and increased gingival exudate

6

(Löe and Holm-Pedersen, 1965, Egelberg 1966, Oliver et al., 1969; Rüdin et al.,

1970); (iv) the clinical signs are not associated with attachment loss, and are therefore,

reversible when the aetiologies are removed.

Gingival diseases that are dependent on the presence of dental plaque for their

initiation fall into two categories; plaque-induced gingivitis that is associated with

local factors, and plaque-induced gingivitis that is affected by local factors and

modified by specific systemic factors found in the host.

1.2.2.1 Plaque-induced gingivitis associated with local factors

Plaque-induced gingivitis is the result of bacterial presence. The aetiology of plaque

bacteria was convincingly demonstrated by classical studies of experimental gingivitis

in humans (Löe et al., 1965; Theilade et al., 1986). These studies have shown that in

healthy individuals, gingivitis can develop when plaque bacteria accumulate,

particularly at the gingival margin, and can be reversed by removing the plaque.

Plaque-induced gingivitis has been shown to be prevalent at all ages in dentate

populations (U. S. Public Health Service, 1965; U. S. Public Health Service, 1972;

Stamm, 1986; U. S. Public Health Service, 1987; Bhat, 1991). This disease is

considered to be the most common form of periodontal disease (Page, 1985).

Clinically this disease is an inflammation of the gingiva. The normal `healthy' firm,

regular contour of the gingiva appears swollen. In light skinned people the normal

pink colour may appear red or blue-red (Mühlemann and Son, 1971; Polson and

Goodson, 1985), however, in dark skinned individuals, this may not be so obvious.

Within 10-20 days clinical signs of inflammation can be detected in most individuals,

although this varies between individuals markedly depending on whether an

individual is resistant or prone to the disease (Van der Weijden et al., 1994).

The clinical changes in the early stages of gingivitis are usually subtle, although the

intensity of the symptoms will vary amongst individuals as well as between sites

within a dentition (Mariotti, 1999). Histologically the symptoms of gingivitis are

quite marked. Many capillary beds are opened as alterations in the vascular network

7

occur. The tissues begin to take on the characteristic swollen appearance as exudative

fluid, proteins and inflammatory cells move into the connective tissues subjacent to

the junctional epithelium. The cellular infiltrate is comprised of mainly lymphocytes,

macrophages and polymorphonuclear leukocytes (PMN). The lymphocytes and

macrophages adhere to the collagen matrix and remain there, whereas the PMN tend

to migrate into the gingival sulcus (Kinane and Lindhe, 1997). Other changes include

proliferation of basal and junctional epithelium, which leads to apical and lateral cell

migration. A marked reduction in the amount of collagen and fibroblasts occurs

(Kinane and Lindhe, 1997). As previously mentioned, this disease is reversible, and

radiographic analysis and examination of individuals with plaque-induced gingivitis

do not reveal loss of supporting tissues (Mariotti, 1999).

1.2.2.2 Plaque-induced gingivitis associated with local factors and modified by

specific systemic factors

It has been shown that gingival disease can be associated with many different factors.

Endocrinotropic gingival diseases

Studies undertaken in the eighteenth century revealed that an exaggerated gingival

response can be seen during pregnancy (Eiselt, 1840; Pinard, 1877). Since these early

reports, many others have followed indicating that sex hormones such as androgens,

oestrogens and progestins can have an effect on the periodontal tissues. These types

of gingival diseases are mainly seen in females at the time of distinct hormonal events,

such as menstruation (menstrual-cycle-associated gingivitis) (Hugoson, 1971) and

pregnancy (pregnancy-associated gingivitis) (Hugoson, 1971; Lbe and Silness, 1963;

Löe, 1965; Arafat, 1974). They can also be seen in both sexes during puberty

(puberty-associated gingivitis) (Mariotti, 1994). It should be noted that although the

sex hormones can modulate the tissues during this disease, plaque micro-organisms

must also be present to produce the gingival response.

8

Drug influenced gingival diseases

Drug-induced gingival enlargement has a number of characteristics and there is a wide

variation in inter-patient and intra-patient patterns (Hassell and Hefti, 1991; Seymour

et al., 1996). It is more often found on the anterior gingiva (Hassell and Hefti, 1991;

Seymour et al., 1996) and there is a higher prevalence of this disease in children

(Esterberg and White, 1945; Rateitschak-Plüss et al., 1983; Hefti et al., 1994). Its

onset is usually within 3 months (Hassell and Hefti, 1991; Hassell, 1981; Seymour,

1991; Seymour and Jacobs, 1992) and enlargement is first observed at the interdental

papilla (Hassell and Hefti, 1991). A change in gingival contour is seen which leads to

modification of its size, and there is a change in colour of the gingiva and an increased

gingival exudate. Clinically bleeding on probing is seen and it is found in the gingiva

with or without bone loss, but it is not associated with attachment loss (Hassell and

Hefti, 1991; Seymour et al., 1996). In this disease, as with all of the gingival diseases,

there is a pronounced inflammatory infiltrate in response to the dental plaque present.

Removal of the bacterial aetiologies can limit the severity of the lesion (Mariotti,

1999).

This disease is associated with anticonvulsants (e. g. phenytoin), immunosuppressants

(e. g. cyclosporin A) and calcium channel blockers (e. g. nifedipine, verapamil,

diltiazem, sodium valproate) (Hassell and Hefti, 1991; Seymour et al., 1996).

Oral-contraceptive associated gingivitis

Use of oral contraceptive agents has been linked to the development of gingival

overgrowth. Several reports have shown the development of this disease in patients

that had previously healthy gingivae (Kaufman, 1969; Lynn, 1969; Sperber, 1969).

The reports indicate that the disease is reversed when the patients discontinue the

contraceptive, or the dose is reduced. However, it has been shown that this disease is

developed in the presence of very little plaque.

9

1.2.2 3 Gingival diseases associated with systemic diseases

Diabetes mellitus-associated gingivitis

A number of studies have been carried out investigating a possible link between

diabetes and diseases of the periodontium (Bacic et al., 1988; Campbell, 1972). It has

also been reported that diabetes mellitus-associated gingivitis seems to be a common

feature of children with poorly controlled type I diabetes. The characteristics

associated with this type of gingivitis are very similar to those of plaque-induced

gingivitis, however, the control of the diabetes is much more important that control of

plaque levels (Cianciola et al., 1982; Gusberti et al., 1983; Ervasti et al., 1985).

Haematologic gingival diseases

One example of these diseases is leukaemia-associated gingivitis. Most of the studies

on this disease type have noted an association between the acute forms of leukaemia

(Lynch and Ship, 1967). The characteristics of the disease are inflammation of the

gingivae which give it a swollen, glazed appearance and the tissues appear spongy and

purple (Dreizen et al., 1984).

Although most of the cases of gingival diseases seen are plaque-induced, non-plaque

associated types of gingivitis do exist. These can be related to bacterial, viral or even

fungal infections. They will not be discussed in this thesis.

1.2.3 Periodontitis

Periodontitis is distinguished from gingivitis by the inflammation which extends to

the periodontal structures, leading to a loss of connective tissue attachment of the

teeth (Ranney, 1991). It has been recognised that there are different forms of

periodontitis, and classifications of the different types is based on the classification

system developed at the 1999 International Workshop for the Classification of

Periodontal Diseases and Conditions.

10

1.2.3.1 Chronic periodontitis

Until the World Workshop on classification in 1999, this form of periodontitis was

referred to as adult periodontitis. It was characterised as a form of periodontitis that

has an onset after the age of 35. The point of contention over the name for this form

of periodontitis comes from the fact that although the prevalence, extent and severity

of the disease increase with age, it is very difficult to determine at what age exactly

the onset of the disease will occur. Although the disease is seen most commonly in

the adult population, it can also be found in children and adolescents. Therefore, for

this reason it has been renamed chronic periodontitis.

Before periodontitis can occur there has to be a pre-existing gingivitis, and hence

bacterial flora in association with the tissues. However, not all cases of gingivitis

progress to periodontitis (Anerud et al., 1979; Listgarten et al., 1985). Periodontitis in

contrast to gingivitis is seen in a much smaller subset of the population. Results from

studies suggest that relatively few subjects in each age group suffer from advanced

periodontal destruction, and that site predilection may be evident in the individuals

affected (Jenkins and Kinane, 1989; Löe et al., 1965; Papapanou et al., 1988).

Clinically, periodontitis is described by a number of features including clinical

attachment loss, alveolar bone loss, periodontal pocketing and gingival inflammation

(Flemmig, 1999). In addition to these features, enlargement and recession of the

gingiva, bleeding on probing and increased mobility, drifting and tooth exfoliation

may occur (Page and Schroeder, 1976).

The recent World Workshop on classification (1999) decided that chronic

periodontitis can be further characterised by the extent and severity of the disease and

can be designated either localised chronic periodontitis or generalised chronic

periodontitis. A guide has been given that disease can be termed localised if <30% of

the sites are affected and generalised if >30% of the sites are affected.

11

1.2.3.2 Aggressive periodontitis

This form of periodontitis was previously referred to as early onset periodontitis. Like

adult periodontitis, there has been controversy over the name early onset periodontitis,

as the disease is not confined to individuals under the chosen age of 35, which is used

for diagnosis. It has therefore, been decided that diagnosis of the disease should not

be based on the age at the which the patient presents his or her symptoms, but by

clinical, radiographic, historical and laboratory findings, as this is a more reliable

method (World Workshop Consensus Report, 1999).

Aggressive periodontal disease is identified by common features which include: (i)

that apart from periodontal disease the patients are fit and healthy; (ii) rapid

attachment loss and bone destruction must be present; and (iii) that there is familial

aggregation. The 1999 World Workshop also decided that the disease, sometimes but

not always, has other characteristics and have therefore, published a list of secondary

features that may be useful in diagnosis. These include: (i) that the amount of

microbial deposits seen are inconsistent with the severity of the tissue destruction; (ii)

that there are elevated proportions of Actinobacillus actin omycetemcoin itans (A.

Actinonryceterncomitans), and in some populations Porphyromonas gingivalis (P.

gingivalis); (iii) there are phagocyte abnormalities; (iv) there may be a hyper-

responsive macrophage phenotype evident, including elevated levels of prostaglandin

E2 (PGE2) and interleukin-1(3 (IL-lß); and (v) that the progression of attachment and

bone loss may be self arresting (World Workshop Consensus Report, 1999).

There are two forms of aggressive periodontal disease; localised aggressive

periodontitis and generalised aggressive periodontitis. Both of these are classified by

the common features listed above, however, the disease entities may be subdivided by

the following features. Localised aggressive periodontitis may have: (i) a

circumpubertal onset; (ii) a robust serum antibody response to infecting micro-

organisms; and (iii) the disease may be localised to a first molar/incisor. There should

also be interproximal attachment loss on at least two permanent teeth, one of which is

a first molar, and involving no more than two teeth, other than first molars and

incisors.

12

Generalised aggressive periodontitis may be diagnosed by the following

characteristics: (i) usually the individuals presenting with this form of the disease are

under 35 years of age. However, as already discussed, age is a disputable factor to use

for diagnostic purposes, because classification will change the diagnosis as the patient

grows older. The time that a patient presents with the disease does not necessarily

mean that it is the age of onset, and therefore, may be older than 35; (ii) that

individuals with this form of the disease have a poor serum antibody response to

infecting agents; (iii) the destruction of attachment and alveolar bone is episodic; and

(iv) there is a generalised interproximal attachment loss affecting at least 3 permanent

teeth, other than first molars and incisors.

1.3 The Microbiology of Periodontal Disease

1.3.1 Introduction

Although there are various environmental factors and host predispositions associated

with the onset of periodontal disease, it is accepted that this disease is associated with

microbial plaque, that is micro-organisms colonising tooth surfaces (Socransky and

Haffajee, 1993).

The oral cavity is made up of a number of distinct habitats, all of which experience

different ecological conditions. These different ecological niches all support different

populations of bacteria, and have a microbial community distinctive to themselves

(Marsh, 1989). This microbial colonisation of the oral cavity originates from birth

and it is thought that the colonising bacteria are acquired from the mother. Evidence

for this theory comes from studies on scrotype tracing, bacteriocin patterns and more

recently by restriction endonuclease mapping (Berkowitz and Jordan, 1975; Kulkarni

et al., 1989).

There are several ways of detecting and sampling oral micro-organisms. The most

commonly used sampling techniques in the detection of bacterial species in

periodontal samples are dental curettes and paper points (Baehni and Guggenheim,

1996). Quantification of oral micro-organisms can be difficult. One of the main

13

methods used in characterisation of the oral microbiota is that of in vitro culturing

methods, and hence, culturing observations (Baehni and Guggenheim, 1996). When

using this approach, the sample is plated out on to non-selective agar, and the

`predominant cultivable flora' detected and identified. Although an effective method,

this approach can be technique-sensitive and very difficult to carry out. In addition it

has been estimated that less than 50% of the total flora from subgingival samples can

be cultivated (Loesche et ei., 1992b). Other molecular biology techniques have been

developed over recent years including immunoassay and DNA probe methods.

Three different types of DNA probe method have been developed; whole-genomic

probes, cloned probes and synthetic oligonucleotide probes (Dewhirst and Paster,

1991). The basis for this method is that DNA segments or sequences of nucleotides of

16S rRNA unique to each bacteria are isolated. DNA probes prepared from a

representative strain of the target micro-organism will hybridise to these sequences,

hence, the detection of these bacteria is due to the binding of the DNA probe to the

complementary strands of nucleic acids in these unique regions (Baehni and

Guggenheim, 1996).

Immunological methods have also been developed to detect different bacteria species

(Tanner et al., 1991). Immunoassays include immunoprecipitation,

immunofluorescence, immunohistochemistry, flow cytometry and Enzyme Linked

Immunosorbent Assay (ELISA). Most of these methods do not identify the bacteria

directly. In general polyclonal or monoclonal antibodies against species-specific

surface antigens can be used to identify target bacteria (Baehni and Guggenheim,

1996).

Research by taxonomists has identified 300-400 bacterial species as being present in

the human oral cavity (Moore and Moore, 1994). Some species are recognised as

being transient, however, most are permanent residents (Darveau et al., 1997). These

have fully adapted themselves to living in the environment offered to them by both the

tooth surface and the surrounding conditions of the oral cavity and gingival epithelium

(Darveau et al., 1997). Most of the time a dynamic equilibrium exists between the

plaque bacteria and the host. Even in health, there are large numbers of micro-

14

organisms present in the oral cavity. The bacteria have adapted themselves to be able

to survive in the environment, and the hosts immune system, both innate and adaptive

arms, normally limit the growth of these micro-organisms. As mentioned, in health

these micro-organisms challenge the host to maintain an effective defence. Under

select conditions that may involve acquisition of certain species, combinations of

species or less than optimal host defence, these bacteria can cause destructive

inflammation (Darveau et al., 1997).

If a tooth is freshly cleaned, a bacterial coating occurs rapidly on its surface (Marsh

and Bradshaw, 1995). Within the first few hours, on the tooth surface a pellicle forms

that consists of proteins and glycoproteins, that are found in saliva and gingival

crevicular fluid (GCF) (Marsh and Bradshaw, 1995). Numerous studies have

indicated the role of the pellicle in providing specific receptors for bacterial

attachment (Duan et al., 1994; Gibbons et al., 1991; Hsu et al., 1994; Scannapieco,

1994; Scannapieco et al., 1989; Scannapieco et al., 1995). The bacterial attachment is

enhanced by the formation of the pellicle in both initial bacterial colonisation, and

also additional bacterial attachment (Skopek and Liljemark, 1994). The term co-

aggregation (Whittaker et al., 1996), is used to describe the phenomenon of two

genetically different bacteria recognising and binding to one another. This is the

mechanism by which the formation of dental plaque biofilm occurs (Whittaker et al.,

1996), and is based on specific interactions of a proteinaceous adhesion produced by

one bacterium and a respective carbohydrate or protein receptor found on the surface

of another bacterium (Darveau et al., 1997). Streptococcus species and

Fusobacterium are thought to play a major role in the supragingival biofilm formation

(Whittaker et al., 1996).

Once the bacterial biofilm has been established, bacterial growth will occur.

Characteristically, biofiims contain areas of high and low bacterial biomass interlaced

with aqueous channels of different size (Costerton et al., 1995). The biofilm is

designed to give each bacterial species its optimal environment for growth and gaining

nutrients. There are two different types of dental plaque biofilms, each of which is

composed of distinctly different species; supragingival plaque, which forms above the

15

gingival margin and subgingival plaque which forms below the margin (Darveau et

al., 1997).

Supragingival plaque resides above the gingival margin and is, therefore, found in the

open space of the oral cavity. Due to its location it experiences constant disruption

due to mastication and salivary flow, which restricts its net accumulation (Sissons et

al., 1995). Supragingival plaque is also exposed to the innate immune mechanisms

found in the saliva of the host. Saliva firstly, flows around the oral cavity creating a

`washing' characteristic. It also contains numerous host defence components

including secretary IgA, lactoferrin, lysosyme, and peroxidases (Shenkels et al.,

1995). Plaque accumulation occurs in the absence of oral hygiene.

Subgingival plaque resides in a more protected location and the space available for

this bacterial growth is limited in periodontally healthy individuals (Darveau et (1l.,

1997). More space can be made however, by reduction of the epithelial cell

attachment levels and by an increase in pocket depth, related to this and the gingival

swelling. GCF is not always advantageous to the host. Although it contains the

innate and adaptive components of the host defence (Cimasoni, 1983), it also contains

nutrients for the biofilm. In terms of the studies carried out in this thesis, subgingival

plaque plays a more pivotal role in periodontal disease and the immune response that

is induced. The following discussion highlights the species of importance in this

disease, the characteristics they possess that make them harmful and their role in the

disease progression. The problem with a discussion of this type when considering

periodontal disease, is that unlike most other diseases, where the presence of some

bacterial species can indicate disease, prevalence does not always correlate with

periodontal disease. It is known that dental plaque is a complex permanent

community and that the presence of large numbers of an indigenous species alone is

not enough to lead to periodontal disease. The balance of the species probably plays a

role, as well as the characteristics of the teeth, the gingival crevice and the host itself

(Liljemark and Bloomquist, 1996).

The bacteria resident in plaque could be advantageous to the host as well as making

the host susceptible to disease. Firstly, it has been suggested that exposure to normal

16

microbial flora (commensal flora) may protect against colonisation by more

pathogenic strains of the same species. Many indigenous microbes are thought to be

closely related to their more virulent relatives and often possess common antigens

(Liljemark and Bloomquist, 1996). Antibodies induced in an individual against a

common antigen may provide a rapid immune response against a more pathogenic

strain should the situation arise.

Evidence for the accepted dogma, that most periodontal diseases are caused by

bacteria in dental plaque (Socransky and Haffajee, 1990; Socransky and Haffajee,

1991; Socransky and Haffajec, 1992), is strong and comes from several sources

including: (i) Studies that show a correlation between most forms of gingivitis and

periodontitis, and accumulated dental plaque; (ii) evidence suggesting that elimination

of plaque micro-organisms leads to clinical improvement; and (iii) in vivo and in vitro

reports indicating that different plaque micro-organisms have been shown to have

virulent properties (Zambon, 1996).

The last 3 decades have seen the emergence of the concept of microbial specificity in

the aetiology of periodontal disease (Loesche, 1976; Socransky, 1977). Until

discussions began as to the relative importance of individual bacterial species within

the dental plaque, it was believed that periodontal disease resulted from the gross

accumulation of dental plaque. This theory, known as the non-specific plaque

hypothesis, suggested that the plaque associated with periodontal disease causes the

disease when the homogeneous mass accumulates to the point of exceeding the host

defences. This hypothesis explained previous clinical experiences when investigators

linked the universal presence of gingival inflammation and periodontal pocket

formation with the presence of abundant plaque (Tanner, 1988). Loesche (1976)

suggested that periodontal disease be considered as non-specific because: (i) the lack

of bacterial invasion. Bacterial invasion has been demonstrated, but it does not

appear, as yet, to be part of an acute phase of disease progression; (ii) their apparent

non-specific nature; (iii) their chronicity; and (iv) their universality. This non-

specific hypothesis was also supported by Löe et al. (1965), who demonstrated that

cessation of oral hygiene is associated with the concominant development of

17

gingivitis, and that resumption of oral hygiene eliminated the accumulated plaque

mass and resolved the marginal gingivitis.

This non-specific theory has been disregarded because the 300-400 bacterial species in

the oral cavity show different characteristics and therefore, must play different roles

(Dahlen, 1993). The non-specific hypothesis is contrasted by that of the specific

plaque hypothesis. This states that the dental plaque micro-organisms isolated from

periodontitis lesions are qualitatively different from those isolated from healthy sites

(Loesche, 1976). Further evidence from cross-sectional and longitudinal studies of

the predominant cultivable microflora, indicated that although periodontal diseases are

polymicrobial infections, only a small number of species are associated with human

periodontal disease (Moore et al., 1983; Theilade, 1986; van Winkelhoff et al., 1988a;

Haffajee & Socransky, 1994).

The concept of bacterial specificity in periodontal disease has been further supported

by clinical observations, by therapeutic effect of control of these bacteria, and by

experimental models of periodontitis in both gnotobiotic and conventionally

maintained animals. However, the continued finding of increased numbers of a range

of species in periodontitis patients, the presence of suspected agents in inactive sites,

the failure of "specific" antibiotics to stop disease progression, and the absence of

these agents in some active sites indicates that the specific plaque hypothesis may be

invalid (Tanner, 1988).

Therefore, while it is accepted that gingival inflammation can be associated with the

build up of non-specific dental plaque micro-organisms, periodontitis may be

associated with a much smaller number of bacterial species. The actual identification

of these important pathogenic species remains elusive, but there are a number of

bacteria which are considered possible candidates.

In an attempt to identify periodontal pathogens the approach has been to use the

criteria known as Koch's postulates. Koch's postulates were originally developed

when Robert Koch was attempting to identify the aetiology of monoinfections such as

the importance of Mycobacterium tuberculii in tuberculosis (Zambon, 1996). They

18

are less useful in the quest for the identification of periodontal pathogens for a number

of reasons, firstly, it is a polymicrobial infection. Further to this, the postulates cannot

be fulfilled for organisms that cannot be grown in pure culture, that demonstrate a

long incubation period, that express pathogenicity first after another infectious agent

has weakened the host immune response and that exhibit a range that is restricted to

humans or to animal species in which the human disease cannot be reproduced (Slots,

1999). Therefore, in recent years, Koch's postulates have been extended by

periodontal research workers, to try and develop a set of postulates especially for the

study of chronic infectious diseases such as periodontitis. As reported by Socransky

(1977), the criteria used for identifying bacteria as being important in the aetiology of

periodontal disease include: (i) compared to low numbers or absence in healthy or

non-progressing sites there is a high number of the micro-organism in the periodontal

lesion. The micro-organism is to be identified at numbers above a critical threshold

prior to initiation of progressive disease, and statistical determination of the micro-

organism as a risk factor; (ii) clinical improvement in the site should be seen on

elimination of the micro-organism; (iii) host responses to the micro-organism should

be seen. Indicators of this are high levels of specific antibody in the serum, saliva or

GCF and a cell mediated immune response; (iv) clinical disease can be correlated with

virulence factors; and (v) demonstration of tissue destruction in the presence of the

micro-organism in appropriate animal models.

A small number of the 300-400 of species that inhabit the oral cavity have been

implicated as periodontal pathogens. These species include; Actinobacillus

actinomycetemcomitans (A. actinomycetemcomitans), Bacteroides forsythus (B.

forsythus), Canipylobacter rectus (C. rectus), Fusobacterium nucleatum (F.

nucleatum), Prevotella intermedia (P. intermedia), Prevotella nigrescens (P.

nigrescens), Porphyromonas gingivalis (P. gingivalis) and treponemes (Zambon,

1996).

Longitudinal studies examining the periodontal pathogens in sites undergoing clinical

attachment loss have provided much of the evidence implicating pathogens in

periodontal disease. In brief, pocket depth, attachment levels, subgingival

temperatures, bleeding on probing, suppuration, redness and plaque were all

19

monitored in a group of 67 patients with no prior evidence of destructive periodontal

disease. In addition to these clinical indices, the levels of 14 subgingival species were

determined in subgingival plaque samples. These included A. actinomycetemcomitans

serotypes a and b, B. forsythus, F. nucleatum ss vincentii, Peptostreptoccus micros, P.

gingivalis, P. intermedia homology groups I and II, Streptococcus interinedius and C.

rectos, also four beneficial species Capnocytophaga ochracea, Streptococcus Banguis

I and II and Veillonella parvula. The results of this particular study confirmed the role

of certain suspected pathogens (Haffajee et al., 1991). However, microbial species

have been shown not to be evenly distributed from subject to subject or even from site

to site, and even more so between different periodontal disease entities. A further

problem connected with the clinical indices is that it is generally impossible to sample

subgingival plaque bacteria at exactly the same time as clinically detectable

connective tissue loss occurs. Therefore, speculation arises as to whether the resulting

micro-organisms are the cause of the connective tissue loss or arrive secondary to it

(Zambon, 1996).

Of the putative periodontal pathogens, stronger evidence exists for some more than

others. The following discussion looks at the micro-organisms considered to be the

more important species in periodontal disease. It is important to stress that these

bacterial species are currently considered as putative periodontal pathogens, however,

whether these pathogens are true aetiologic agents remains elusive.

1.3.2 Porplryromonas gingivalis

Members of the Porphyromonas gingivalis species have the characteristics of being

non-motile, asaccharolytic, Gram negative and obligately anaerobic coccobacilli

which exhibit smooth, raised colonies. The morphology of the colonies may vary

from smooth to cauliflower shaped on blood agar plates. The normal habitat of P.

gingivalis is the oral cavity and most likely the gingival sulcus (Olsen et al., 1999).

The organism is rarely found in supragingival plaque or outside the oral cavity (van

Winkelhoff et al., 1988a).

20

These species show excellent properties for evading the host immune responses due to

the production of enzymes, proteins and end products of their metabolism, which are

active against a large number of host proteins (Holt et al., 1999). Many strains of P.

gingivalis possess an external capsule. A strong correlation has been identified

between the extent of encapsulation and several biological functions that have a

significant effect on the biological properties of the bacteria (Holt et al., 1999).

Studies have suggested that there is a decreased tendency of the encapsulated strains

to be phagocytosed, an important factor for host response evasion (Schifferle et al.,

1993, Sundqvist et al., 1991). P. gingivalis has been shown to differ from other Gram

negative bacteria, in that it does not have the capacity to directly stimulate the

production of E-selectin by human endothelial cells, thus, hindering the migration of

leukocytes into the extravascular compartment (Darveau et al., 1995). Further to this,

lipopolysaccharide (LPS) of P. gingivalis has also been shown to bind LPS binding

protein (LBP) very poorly, leading to poor myeloid cell activation (Cunningham et al.,

1996). Both of these strategies by this micro-organism hinder the protective

characteristics of the innate host response. In addition, the cysteine proteases of P.

gingivalis appear to be of great physiological importance, and will be discussed later

in section 1.6.7.

P. gingivalis is found rarely in health, however is often found in disease, particularly

in periodontitis (Slots and Listgarten, 1988; van Winklelhoff et al., 1988a; Moore et

al., 1991). It has also been shown that there is a higher load of P. gingivalis in active

lesions compared with inactive lesions (Dzink et al., 1988; Walker and Gordon,

1990). Evidence from studies investigating the effect of therapy on disease, indicate

that levels of P. gingivalis remain high in sites that have responded poorly to therapy

(van Winkelhoff et al., 1988b; Choi et al., 1990). High proportions of P. gingivalis

were found in sites following breakdown of treatment, (Choi et al., 1990) indicating

that this species may play an important role in the pathogenesis of disease recurrence.

1.3.3 Actinobacillus actinonryceteincomitans

Actinobacillus is a member of the family of Pasteurellaecae. This family are a group

of nonenteric, fermenting Gram negative rods. The term Àctinobacillus' refers to the

21

internal star shaped morphology of its bacterial colonies on solid media, and to the

short rod or bacillary shape of individual cells. Many of the family are found in

animals, however, only A. actinoniycetemcomitans is routinely cultured from humans

(Lally et al., 1996). A. actinomycetenicoinitans has been associated with a number of

disease entities in man including; infective endocarditis, brain abscesses,

osteomyelitis, subcutaneous abscesses and several forms of periodontal disease (Block

et al., 1973; Page and King, 1966; Zambon et al., 1983; Zambon, 1985).

Most periodontally healthy adults show low or no detectable levels of subgingival A.

actinoniycetemcoin itans (Alaluusua and Asikainen, 1988; Gmür and Guggenheim,

1994). There are five known scrotypes of A. actinoinycetemcomitans, a-e, and studies

have indicated that serotype c seems to be found predominantly in health and that

serotype b is found more frequently in patients with periodontal disease (Asikainen et

al., 1991).

A. actinonrycetemcomitans was first described to be found in increased numbers in

patients with localised aggressive periodontitis (Slots et al., 1980; Zambon, 1985).

More recently, studies on localised aggressive periodontitis have shown A.

actinomycetemcomitans to be higher in numbers in periodontal lesions when

compared with non-diseased sites (van Winkelhoff et al., 1994), and in active sites

compared to inactive sites (Haffajee et al., 1984; Mandell, 1984). Although, A.

actinomycetemcomitans has been linked predominantly with aggressive periodontitis,

it is still implicated in chronic periodontitis (Gmür and Guggenheim, 1990).

Further evidence of a pathogenic role for A. actinotnycetemcomitans comes from

studies investigating the effects of periodontal treatment on the microbiological

parameters. Studies have reported a correlation between successful treatment and

suppression of A. actinomycetemcomitans, below detectable levels and unsuccessful

treatment and a failure to suppress numbers of this micro-organism (Christersson et

al., 1985; Mandell et al., 1986). Given the virulence of A. actinomycetemcomitans a

number of authors have suggested that elimination of this micro-organism should be

the goal of any periodontal therapy. Failure to eliminate this bacteria can lead to

continued attachment loss or reduced healing (Zambon, 1996).

22

This species produces a number of virulence factors, such as cell surface

carbohydrates and leukotoxin. Leukotoxin has drawn much interest and may play an

important role in periodontal disease. It will not be covered here as a more

comprehensive discussion of the topic can be found in section 1.6.6.

1.3.4 Prevotella intermedia

Prevotella intermedia is a black pigmented, short, round-ended anaerobic rod. It is

thought that P. interinedia has a number of virulence properties also exhibited by P.

gingivalis. Recently two species of P. intermedia that show identical phenotypic traits

have been separated into two species; P. intermedia and P. nigrescens (Shah and

Gharbia, 1992).

P. intermedia has been associated with many forms of periodontal disease (Slots,

1982; Loesche et al., 1982; Slots and Listgarten, 1988; Williams et al., 1985). Levels

of P. intermedia have been shown to be elevated in certain forms of periodontitis

(Dzink et al., 1983, Moore et al., 1985; Tanner et al., 1979). Both the frequency of

detection and levels of P. intermedia are increased in active periodontitis lesions

(Slots et al., 1986; Dzink et al., 1988; Moore et al., 1991), and it has been detected

attached to epithelial cells in increased numbers at diseased sites (Dzink et al., 1989;

Dibart et al., 1998). P. intermedia is a likely candidate for a periodontopathic

organism, however, it does not appear to be as prevalent or pathogenic as P.

gingivalis.

1.3.5 Bacteroidesforsythus

Bacteroides forsythus is a Gram negative, anaerobic, spindle shaped, highly

pleomorphic rod. The last decade has seen the utilisation of new molecular biology

techniques, such as detection of bacteria in plaque samples using immunofluorescence

with polyclonal and monoclonal antibodies (Lai et al., 1987, Gmür, 1988) and DNA

probes (Moncla et al., 1991). This work has led to suggestions that B. forsythus is a

putative periodontal pathogen. Until these studies, B. forsythus was rarely reported to

be one of the more predominant organisms in periodontal lesions. This was probably

23

due to the difficulties that occur when trying to culture B. forsythus. Culture of B.

forsythus in vitro has many specific requirements (Wyss et al., 1989) and often

requires 7-14 days for minute colonies to develop. Growth of this organism in vitro

has been shown to be enhanced by co-cultivation with F. nucleatum, which it has been

shown to commonly occur in subgingival sites (Socransky et al., 1988). B . forsythus

has a requirement for N-acetylmuramic acid, and inclusion of this in media enhances

growth of this micro-organism (Wyss et al., 1989).

B. forsythus has been detected in the supragingival plaque of healthy patients (Gmür

and Guggenheim, 1994). However, studies have shown B. forsythus to be at higher

proportions in patients with gingivitis and periodontitis, than healthy subjects (Lai et

al., 1987; Gmür, 1988; Moncla et al., 1991). Dzink et al. (1988) showed B. forsythus

to be at increased proportions and with increased frequency in active sites compared

to quiescent sites of periodontitis patients, and Lai et al. (1987) also showed increased

numbers of the micro-organism in sites with periodontal breakdown.

1.3.6 Spirochaetes

Of the oral spirochaetes, the Treponema genera appears to be regarded as one of the

more important. There are currently four named human species: T denticola; T.

vincentii; T. pectinovarum; and T. socranskii. The organism T. denticola has been

consistently reported to be associated with periodontal disease (Simonson et al., 1988;

Haapasalo et al., 1992; Zambon, 1996), and is the cultivable spirochaete which

predominates in most patients with chronic periodontal disease (Moter et al., 1998).

Spirochaetes are not usually detected in healthy sites, but reach high levels in sites of

disease. It is thought they average 30-50% of the microscopic counts in subgingival

plaque samples recovered from periodontal pockets (Listgarten and Hellden, 1978).

However, it has been difficult to ascertain whether the oral treponemes are present in

large numbers in subgingival plaque from periodontitis sites and discover whether

they are aetiologically relevant. Moter and colleagues (1998) recently suggested that

the environment of the periodontal pocket is a very suitable niche, with its anaerobic

environment and plentiful supply of host and microbial molecules for enriched growth

of these species.

24

Spirochaete numbers are generally increased in patients with chronic periodontitis and

can be found in very high prevalence (60-100%) (Loesche et al., 1985; Savitt and

Socransky, 1984; Riviere et al., 1992; Riviere et al., 1995). They have also been

linked with increasing pocket depth, bleeding on probing and bone loss (Omar et al.,

1991; Tanner et al., 1984; Savitt and Socransky, 1984).

A number of potential virulence factors have been reported for T. denticola and

following the trend seen for other putative periodontal pathogens, potential significant

antigens have been suggested. These include: a periplasmic flagella, which provides

motility (Boehringer et al., 1986); and two proteases (Ohta et al., 1986; Grenier et al.,

1990). The latter could contribute to the pathogenesis of periodontal disease, through

their activity on other bacteria and host proteins (Grenier, 1996). A 53 kDa porin

(Olsen et al., 1984; Haapasolo et al., 1992; Egli et al., 1993); some major outer-

membrane proteins, which appear to function in attachment of the micro-organism to

various host cells (Weinberg et al., 1990); and cystalysin, which enzymatically

produces H2S and acts an hemolysin (Chu et al., 1997). Difficulties encountered

when working with these often uncultivable species, have contributed to minimal data

detailing the host immune responses to the oral treponemes.

Further evidence indicating the importance of T. denticola is given in a study by

Simonson et al. (1992) which shows a decrease in the proportion of T. denticola

following successful treatment, however, MacFarlane et al. (1988) reported that

spirochaetes are poor predictors of future disease activity.

1.3.7 Microbial composition associated with health

Further studies of the microbial composition of a healthy gingivae could provide

important information regarding the distribution and numbers of bacteria present

between health and disease. This could also lead to a clearer picture of micro-

organisms that colonise and play a more important role in disease progression.

The nature of periodontal disease makes comparison with healthy gingiva difficult as

mentioned earlier. Subgingival plaque resides in more protected locations and the

25

space available for bacterial growth is limited in periodontally healthy individuals

(Darveau et al., 1997). Much of the data regarding health has come from evaluation

of healthy sites that have been selected as controls in other studies. The type of

studies that this kind of data is obtained from includes; the starting point of

experimental gingivitis (Listgarten et al., 1975; Moore et al., 1982; Syed and Loesehe,

1978; Theilade et al., 1966) and healthy sites in periodontally diseased subjects

(Moore et at., 1987; Moore and Moore, 1994; Tanner et at., 1979). The literature

suggests the bacterial load associated with health in young individuals is relatively

low and the micro-organisms present are predominantly Gram positive, streptococci

and actinomyces. Approximately 15% of the microbiota isolated seem to be Gram

negative rod species (Moore et al., 1982; Syed and Loesche, 1978; Tanner et at.,

1996). The situation appears to be a little different in older individuals. It has been

reported that in healthy individuals, with no signs of previous gingivitis, up to 45% of

the microbiota isolated from them appeared to be Gram negative bacteria (Newman et

al., 1978; Slots, 1977). Microbiota isolated from healthy sites of adult patients with

periodontitis were also identified as being from a similar range of Gram negative

microflora (Moore and Moore, 1994; Tanner et at., 1979). The results discussed

above suggest that colonisation of healthy sites with Gram negative bacteria occurs,

however, it does not appear to result in disease alone.

1.3.8 Microbial composition associated with gingivitis

Although it is not the only predisposing factor for disease, colonisation by Gram

negative bacteria is necessary for the onset and progression of disease. Many studies,

both cross-sectional and longitudinal, have reported that gingivitis is associated with a

change in the balance of Gram negative and Gram positive organisms and an

increased microbial load. The microbial load has been suggested to change from

approximately 102-103 in health to 104-106 in gingivitis and that there is an increase of

15-50% of Gram negative organisms isolated from gingivitis versus health (Gmür et

al., 1989; Lai et al., 1987; Listgarten et al., 1975; Moore et al., 1987; Moore et al.,

1984; Moore et al., 1982; Preus et al., 1995; Riviere et al., 1996; Slots et al., 1978;

Syed and Loesche, 1978; Tanner et al., 1996; Theilade et al., 1966).

26

1.3.9 Microbial composition associated with periodontitis

A further increase in microbial load is seen in periodontitis compared to health and

gingivitis, and the load is reported to be in the range of 105-108 micro-organisms

(Darveau et al., 1997). A significant proportion of periodontal sites harbour

periodontal pathogens at a high prevalence, and this prevalence is seen to increase in

deep periodontal pockets (Wolff et al., 1993). All of the above micro-organisms are

found at high prevalence in disease sites of patients with periodontal disease.

1.4 The Host Immune Response

1.4.1 Introduction

The immune system can be divided into two broad categories; the innate response and

the adaptive immune response. Innate immunity is present from birth and does not

change or adapt in response to a particular antigen. The adaptive or acquired immune

response is specific and is acquired during the lifetime of an individual. The

following discussion highlights the role of these two arms of the immune response in

the onset and progression of periodontal disease.

1.4.2 The innate immune system

The innate immune response was first described by the Russian immunologist Elie

Metchnikoff (1905). Metchnikoff discovered that many micro-organisms could be

engulfed and digested by phagocytic cells called macrophages. This non-specific

defence against infection is inborn or innate since its action does not depend upon

prior exposure to a particular antigen (Janeway and Travers, 1994).

The innate immune response is important as it is the first line of defence against

infection. It can be divided into two clear phases. The first phase operates

immediately, is constantly present and requires no induction. Physical barriers such as

the skin and mucous membranes represent this component, which infectious agents

must breach to gain access to the host. The washing action of fluids such as saliva,

27

tears, urine and GCF keeps mucosal surfaces clear of invading organisms. These

secretions also contain bacteriocidal agents. The intact epithelial barrier of the

gingival, sulcular and junctional epithelium usually prevents bacterial invasion of the

periodontal tissues. The epithelial cell wall, secretes proteins and fatty acids which

are toxic to many microbes. Salivary secretions provide a continuous flushing of the

oral cavity as well as providing a continuing supply of agglutinins and antibodies. In

addition, the GCF flushes the gingival sulcus and delivers all the components of the

serum. Most epithelia of the body are associated with a normal flora of non-

pathogenic bacteria which compete with pathogenic micro-organisms for nutrients and

attachment sites on cells. The normal flora can also produce anti-microbial

substances that prevent colonisation by other bacteria.

All of these physical and chemical barriers to infection may be considered to be part

of the innate immune system since they are non-specific and do not become more

effective after exposure to a particular pathogen. However, barriers can be breached.

All surfaces of the host can be damaged by cuts, grazes and burns which allow

pathogens entry, and some bacteria can break down the barriers with enzymes. Thus,

a second line of defence is necessary.

Within seconds of a breach in the epithelial barrier, the second phase of the innate

immune response is initiated. This is induced but still non-specific for the particular

invading organisms, and does not generate immunological memory. One of the first

signs of this phase of the immune response is inflammation. An inflammatory

response can be triggered directly by pathogens, especially bacteria. The term itself is

purely descriptive and was originally defined by the four Latin words dolor, rubor,

calor and tumor, meaning pain, redness, heat and swelling. The quality of the host

inflammatory response is critical to the disease process, although its purpose is

protection and prevention of bacterial invasion, it can also be detrimental. It may also

be an ineffective, chronic frustrated response which actually causes much of the tissue

damage that occurs in periodontal disease. As the inflammatory immune response is

non-specific and is induced against many different infectious agents all over the body,

it seems reasonable to assume that it is not defective. Periodontitis patients, are not

28

typically susceptible to infections at other sites and hence it can be assumed that the

inflammatory immune response that they induce, is effective.

1.4.3 The inflammatory response

The effects that occur due to the induction of an inflammatory response result from

dilation of the local blood vessels. These vessels also become permeable and show

expression of adhesion molecules which capture passing leukocytes. The reaction is

accelerated by the first cells to arrive. These cells release molecules or inflammatory

mediators which recruit more cells and which further enhance the reaction. The

inflammatory response is mediated by an array of molecules including complement

components, the contents of mast cell and basophil granules, and molecules secreted

by macrophages. There is strong evidence that inflammatory mediators play a role in

periodontal disease. In the periodontium, inflammatory mediators such as cytokines,

proteinases, thrombin, histamine and prostaglandins are evident. These are produced

and released from the activated infiltrating cells, plasma cells, resident fibroblasts and

other connective tissue cells, and from activated complement and other plasma

proteins (Page, 1991). They are produced in the blood plasma by the complement

cascade and kinin system. Monocytes from individuals susceptible to or suffering

from severe periodontitis, produce elevated amounts of mediators (Garrison and

Nichols, 1989). Mediators are present within the inflamed gingiva and in the GCF of

diseased sites in high concentrations. Concentrations of inflammatory mediators

usually decrease following successful periodontal therapy (Offenbacher et al., 1993).

1.4.4 Complement

Complement is a collective term for a series of about 20 serum proteins that circulate

in the body fluids in an inactive form. When the first component of a complement

cascade is activated, a complex reaction is set in motion. The reaction products of

each step in the sequence activate the next component in the cascade. A complement

cascade can occur due to both innate and adaptive immune responses. The final

products of a complement cascade are known as the membrane attack complex. This

is a cylindrical assembly of molecules that are inserted into the cell wall of the

29

microbe in question. The insertion of this complex, opens up a pore through which

sodium ions and then water flood into the cell, hence causing death (Janeway and

Travers, 1994).

The alternative pathway of the complement cascade can occur in the absence of

specific antibodies when directly triggered by a microbial surface. In this way the

same antimicrobial complement actions take place as seen in the classical pathway,

but without the delay of 5-7 days that it takes to induce a specific antibody response.

The necessary characteristics of a pathogens cell surface that allow direct triggering of

the complement cascade are unknown. Endotoxins such as LPS from cell walls of

Gram negative bacteria have been suggested as a possible trigger (Eley and Cox,

1998).

The protein C3 is abundant in the plasma and C3b is constantly produced by

spontaneous cleavage. Although most of this will be inactivated by hydrolysis, some

will bind covalently to the surface of host cells and pathogens. Binding of C3b leads

to the binding of Factor B and the initiation of the cascade. The C3b that binds to the

surface of host cells will not generally lead to the initiation of further activation steps.

Cell surface proteins such as decay accelerating factor, membrane co-factor protein

and the plasma protein factor H, prevent further activation.

Not only is the membrane attack complex of importance, but in addition the small

fragments of C5 and C3, called C5a and C3a are peptide mediators of inflammation.

These fragments play a role early on and mediate local inflammatory responses,

recruiting leukocytes and protein to the site of infection. They also lead to the

accumulation of fluid. C3a and C5a are also able to attach to receptor sites on mast

and inflammatory cells. They bring about the release of histamine and other

substances from mast cells and prostaglandins from inflammatory cells. These

complement fragments are chemotactic for PMN. They also aid phagocytosis by

attaching the antigen to the phagocyte through the C3 receptor on the surface of PMN,

monocytes and macrophages (Rietschel et al., 1984). Complement activation in the

gingival crevice could provide the chemotactic stimulus for leukocyte recruitment and

the release of leukocyte lysosomal enzymes, further increasing epithelial permeability.

30

Complement activation in the tissues itself can lead to increased vascular permeability

and kinin activity (Nisengard, 1977). Complement studies of GCF from patients with

periodontal disease suggest that complement is activated by the alternate pathway

(Schenkein et al., 1976). Although studies have detected the presence of some

complement factors in both periodontal patients and in subjects taking part in

experimental gingivitis studies (Pattres et al., 1989), none of these factors appear to

have any diagnostic value (Eley and Cox, 1998).

1.4.5 Prostaglandins

Prostaglandins play an important role in inflammation. PGE2 particularly exhibits a

broad range of proinflammatory effects and these effects are enhanced by synergism

with other inflammatory mediators (Williams and Peck, 1977). In vitro studies have

indicated that prostaglandins have numerous effects in periodontal disease. They

mediate bone resorption (Klein and Raisz, 1970; Goldhaber, 1971; Goldhaber et al.,

1973; Gomes et al., 1976). In terms of the inflammatory response, the production of

prostaglandins from gingival tissues increases with inflammation (El Attar, 1976). In

addition human monocytes have been shown to produce PGE2 when stimulated with

LPS from various periodontal pathogens (Garrison et al., 1988). Gingival and

periodontal ligament fibroblasts also secrete PGE2 in response to both IL-lß and in

culture to media conditioned with LPS (Richards and Rutherford, 1988).

1.4.6 Cytokines in inflammation

The term "cytokine" is derived from the Greek word kytos meaning cell and kinesis

meaning movement. Cytokines are small soluble proteins that make up a large and

diverse group of molecules. They have a vast range of potent biological functions.

Cytokines were first discovered in the context of research into the mechanisms of

pyrogenesis and cellular immunology and the first cytokines to be discovered were IL-

1 and IL-2. The number of cytokines discovered over the last 20 years now exceeds

100. Cytokines are referred to as mediators because they are molecules which direct

and regulate inflammation and wound healing. As a rule, the synthesis of cytokines is

inducible, although some factors are known to be produced constitutively.

31

Cytokine regulation is achieved through several mechanisms. Control can take place

at the gene activation level, during secretion and circulation and at the cell receptor

level (Bendtzen, 1994). Although good control mechanisms exist, cytokine

production is usually short lived and self limiting (Howells, 1995). In addition, many

cytokine receptors exist in soluble forms, which may be cleaved from the target cells

allowing them to bind and neutralise cytokines extracellularly (Bendtzen, 1994).

Cytokines can be broadly split into two groups, those involved in inflammatory and

immune reactions, and those involved in tissue growth and repair. The first group can

be further sub-divided into; the interleukins that transmit information between

leukocytes, chemokines or chemotactic cytokines that are involved in cell recruitment,

and interferons which are involved in influencing lymphocyte activity.

1.4.6.1 Interleukin-1

In humans, there are two distinct gene products that have been cloned IL-la and IL-

I P. These molecules are distinct but structurally related (Tatakis, 1993), they have

similar proinflammatory properties, although IL-I 3 is more potent (Stashenko et al.,

1987). Sources of IL-1 production include macrophages, monocytes, lymphocytes,

vascular cells, brain cells, skin cells, fibroblasts (Alexander and Damoulis, 1994),

keratinocytes (Luger et al., 1981), endothelial cells (Miossec et al., 1986) and

osteoblasts (Hanazawa et al., 1987). IL-I is a multifunctional cytokine and is one of

the key mediators of the body's response to microbial invasion, inflammation,

immunologic reactions and tissue injury (Alexander and Damoulis, 1994).

IL-1 is a major mediator in periodontal disease. It has been found in the GCF of

patients with the disease (Reinhardt et al., 1993) and also both as a protein (Jandinski

et al., 1991) and mRNA transcripts (Matsuki et al., 1993) in inflamed gingival tissues.

Production of IL-I is induced by LPS (Manthey and Vogel, 1994) and other bacterial

components, including cell surface carbohydrates (Takada et al., 1993) and porins

(Galdiero et al., 1993).

32

IL-1 is known to stimulate the proliferation of keratinocytes, fibroblasts and

endothelial cells, and also to enhance fibroblast synthesis of type I pro-collagen,

collagenase, hyaluronate, fibronectin and PGE2. From these functions it is obvious

that IL-I is a critical component in the homeostasis of the periodontal tissues (Okada

and Murakami, 1998). However, when the inflammatory response becomes chronic

and there is unrestricted production of IL-I, it can become harmful and promote tissue

damage.


IL-6 is a multifunctional cytokine that regulates immune responses, acute phase

reactions and haematopoiesis (Hirano, 1991). IL-6 is produced by macrophages,

fibroblasts, lymphocytes and endothelial cells. The production is induced by IL-1,

tumour necrosis factor-a (TNF-a) and interferon-y (IFN-y) (Shalaby et al., 1989).

The biological effects of this cytokine are similar and overlap with the properties of

other cytokines including IL-I and TNF-a.

There have been reports of the presence of IL-6 in the periodontal tissues (Kamagata

et al., 1989; Barthold and Haynes, 1991; Shimizu et al., 1992), often at levels higher

than those found in healthy controls. IL-6 is thought to be an essential cytokine for

the terminal differentiation of activated B cells and may play a role in the induction of

the elevated B-cell response seen in the gingival tissues of patients with chronic

periodontitis (Fujihashi et al., 1993).


There are 12 members in the human IL-8 family (Skerka et al., 1993). IL-8 falls into

the category of cytokines known as the chemokines. The chemokines are divided into

two broad groups, a and P. They are distinguished by minor differences in structure

and activation of different cell types. IL-8 is an a-chemokine and this group promotes

the migration of PMN. IL-8 is a chemotactic factor for PMN, inducing them to leave

the bloodstream and migrate into the surrounding tissues, in this case, the gingival

33

tissues. This process occurs in two stages. Firstly, IL-8 arrests the circulation of the

cell. Subsequently it directs the migration of the cell along a gradient that increases in

concentration towards the site of infection.

IL-8 is produced by monocytes, macrophages, fibroblasts and keratinocytes. The

production of mature protein and expression of IL-8 mRNA was demonstrated by

various cell types, including neutrophils, particularly following phagocytosis (Skerka

et al., 1993).

Takashiba et al. (1992), reported that human gingival fibroblasts expressed IL-8

mRNA when stimulated with IL-1(x and IL-1p. In addition, these cells also produced

biologically detectable IL-8 when stimulated with IL-1(3 or TNF-a. This study also

showed an increase in chemotactic activity in the conditioned culture medium which

was in parallel with IL-8 mRNA expression. It is hardly surprising that IL-8 should be

found in the tissues of the periodontium since PMN are the first cells to arrive at the

site of inflammation and infection.

1.4.6.4 Tumour necrosis factor-a

TNF-a and TNF-ß are proinflammatory cytokines, produced mainly by macrophages

and lymphocytes respectively (Manogue et al., 1991). TNF-a and TNF-ß show

considerable sequence homology and bind to the same receptor (Aggarwal et al.,

1985). TNF-a is produced by macrophages in response to the recognition of

microbial constituents, particularly LPS (Beutler et al., 1985), and the local effects of

TNF-a are extremely striking. TNF-a primarily initiates the inflammatory response.

It increases the vascular diameter of the local blood vessels, leading to increased blood

flow and vascular permeability and local accumulation of serum. TNF-a also has an

effect on the endothelium, inducing the expression of adhesion molecules that bind to

the surface of circulating monocytes and PMN. This greatly enhances the rate at

which these cells migrate across capillary walls into the tissue. TNF has been

reported to have similar effects to IL-1 and IL-6 (Manogue et al., 1991).

34

Matsuki et a!. (1992), reported that TNF-a mRNA was abundant in macrophages and

T cells in the gingival tissues of patients with moderate to severe periodontitis.

1.4.7 Adhesion molecules

One of the effects of TNF-a is to act on the endothelium to induce the expression of

adhesion molecules that bind to the surface of circulating monocytes and PMN. This

binding increases the rate at which these cells migrate from the small blood vessels

into the tissues, This process is called extravasation. It is now evident that adhesive

interactions control inflammatory/immune reactions from the initial stages through to

the end stage "effector" functions of host-response cells (Crawford, 1994).

Leukocyte adhesion deficiency (LAD), gives evidence of the beneficial role of cell

adhesion molecules in inflammatory diseases. Patients suffering from this disease

have an abnormal gene that codes for the family of adhesion molecules called the

leukocyte integrins. Patients with this abnormal gene have a dramatically increased

susceptibility to infections and suffer from a very severe form of periodontal disease

that can lead to premature loss of deciduous and permanent teeth (Crawford, 1994).

The main classes of adhesion molecules known to have a role in inflammation and

immunity are the selectins, integrins, members of the immunoglobulin superfamily

and some mucin-like molecules. All of these groups share similar properties. They

are all complex proteins that interact via heterophilic binding with one or more

ligands on other cells or the extracellular matrix. They are all involved in signalling

events across the cell membrane, are regulated by cytokines and their expression and

activity is localised at the site of inflammatory lesions.

1.4.7.1 The selectins

Selectins are cell surface molecules and all members of this group have a similar core

structure but they differ in the lectin-like domain in their extracellular portion.

Lectins are molecules that bind to specific sugar groups, and each selectin binds to a

cell surface carbohydrate molecule. The selectins have therefore, so been named

35

based on their selective distribution and lectin arninoterminal domain (Bevilaqua,

1989; Bevilaqua et al., 1991). There are 3 selectins described to date; E-selectin, P-

selectin and L-selectin. Selectins are particularly important for leukocyte homing to

specific tissues, and can be expressed on either leukocytes (L-selectin) or on vascular

endothelium (P- and E-selectins). Selectin expression is induced by C5a, histamine

and TNF-a and these selectins can be considered as the initial binding molecules.

They mediate leukocyte rolling on the endothelial cells, thus allowing the leukocyte to

be exposed to the same soluble factors as the endothelium. They are expressed

relatively early in the inflammatory process i. e. 3-6 hours after exposure to antigen in

vitro (R&D Systems, 1994).

E-selectin was originally described as an activation antigen on endothelial cells in

inflamed skin (Cotran et al., 1986). It is expressed by endothelial cells at the site of

PMN infiltration (Munro et al., 1991), and constitutively by large blood vessels (Page

et al., 1992a). In the periodontal setting it has been shown that E-selectin is expressed

very early on in the inflammatory process (Moughal, 1992) and is drastically down-

regulated in the more established lesion (Gemmel et al., 1994). It shows high affinity

for PMN (Janeway and Travers, 1994) which explains its elevated expression in the

inflammatory response when PMN numbers are high and decreased expression later

on in the immune response when PMN numbers have declined (Kinane and Lindhe,

1997; Lappin et al., 1999).

P-selectin is only expressed on the cell surface after activation with thrombin,

histamine or phorbol esters (Larsen et al., 1990; Toothill et al., 1990). P-selectin is

contained in Wiebel-palade bodies of endothelial cells and a-granules of platelets. It

is released during clotting, and at times of platelet activation and mediating adhesion

between leukocytes and platelets (Freemont, 1998).

L-selectin is the only selectin to be expressed constitutively at the cell surface. It is

expressed by some lymphocytes, PMN, monocytes and other myeloid cells (Stoolman,

1989). L selectin has been termed the "homing receptor" and is involved in

trafficking PMN into inflammatory lesions.

36

E-, P- and L-selectin all have the common ligand the Siayl-Lewisx moiety. This is

found in the sulphated form as part of the structures of mucin-like vascular addressins:

CD34; Glycosylated Cellular Adhesion Molecule-1 (GlyCAM-1); and Mucosal

Addressin Cellular Adhesion Molecule-1 (MadCAM-1) molecules. Each integrin also

has other ligands (see Table 1.1) (Janeway and Travers, 1994; McMurray, 1996).

1.4.7.2 The integrins

The integrins comprise an extensive family of adhesion molecules and consist of a

large a chain paired non-covalently with a smaller ß chain. There are several

subfamilies of integrins which are defined by their 1 chain. However, this concept is

controversial since it is difficult to group integrins by individual a chains associated

with one ß chain, as one a chain can associate itself with more than one ß chain

(Crawford and Watanabe, 1994). Despite this, 3 distinct families of integrins have

arisen based on (3l, (32 and ß3 chains. The ß1 family, are involved in adhesion to

connective tissue macromolecules such as fibronectin, laminin and collagens. The ß2

family are involved in cell/cell interactions in inflammation and immunological

reactions. The ß3 family, are involved in binding to vascular ligands such as

fibrinogen, van Willebrand factor, thrombospondin and vitronectin (Crawford and

Watanabe, 1994) and are recognised as playing an important role in inflammation.

al ß2 (LFA-1) was the first integrin to be characterised and is expressed on all

leukocytes, with the exception of some macrophages. Ligands of a1ß2 include

Intercellular Adhesion Molecule I (ICAM-1) (Marlin and Springer, 1987), ICAM-2

(Diamond et al., 1990) and ICAM-3 (Vaseuax et al., 1992). LFA-1 is thought to be

the most important adhesion molecule for lymphocyte activation, as antibodies to

LFA-l inhibit the activation of naive and effector T cells. a2ß2 (Mac-1), is expressed

by PMN, monocytes and macrophages and a3ß2 (gp150,95) is mainly expressed by

monocytes and macrophages, and at a lower density on PMN.

37

1.4.7.3 The immunoglobulin superfamily

The immunoglobulin superfamily has a distinguishing feature from other adhesion

molecules, they have a domain that is a compact region of the polypeptide chain

enclosed by a disulphide bond. Members of this family include ICAM-1, ICAM-2,

ICAM-3, LFA-3, and VCAM-1 (Simmons et al., 1988; Staunton et al., 1989; Osborn

et al., 1989).

ICAM-l is expressed on vascular endothelium, thymic epithelial cells, fibroblasts,

macrophages and germinal centre dendritic cells in lymphoid tissue, epithelial cells in

the mucosa of tonsils (Dustin et al., 1986) and Langerhans cells (De Panfilis et al.,

1990). ICAM-1 plays an important role in the inflammatory response, by being

induced on endothelial cells and encouraging the migration of leukocytes into infected

tissues.

ICAM-2 binds to LFA-1 but not Mac-1 (Diamond et al., 1990). It is thought to be

more important in re-circulation of lymphocytes, rather than in extravasation and their

localisation into sites of infection. The reasoning behind this theory is that ICAM-2 is

the major ligand for LFA-1 on resting endothelium (Defougerolles et al., 1991)

ICAM-3 is expressed on unstimulated peripheral blood lymphocytes, monocytes and

PMN (DeFougerolles and Springer, 1992; Fawcett et al., 1992). ICAM-3 is

homologous to ICAM

V-CAM-I is another adhesion molecule that is involved in the extravasation process

of leukocytes into sites of infected tissue. During inflammation at peripheral sites,

postcapillary venules express V-CAM-I in response to cytokines such as IL-I and IL-

8 (Albeda et al., 1994). As the periodontal lesion progresses there is increased

expression of VCAM-1 and ICAM-1, which favours the migration of leukocytes

expressing ligands for these molecules i. e. Very Late Antigen-4 (VLA-4/ (X4ß1

integrin) and LFA-I integrin respectively. VLA-4 is specific for T and B lymphocytes

and is upregulated on these cells when in contact with activated endothelium

38

(McMurray, 1996), hence the increased proportions of lymphocytes in the more

established lesion (Kinane and Lindhe, 1997; Lappin et al., 1999).

Mucosal Addressin Cellular Adhesion Molecule-I (MAdCAM-1) has recently been

identified on the post-capillary venules of gut mucosa and high endothelial venules of

lymphoid tissue and dendritic cells. It is considered to be a specific adhesion

molecule for the mucosal tissue and is an important controlling factor determining

whether a mucosal or non-mucosal immune reaction occurs due to the cell type it

recruits (Briskin et al., 1997). MAdCAM-1 has been shown to have strong homology

with VCAM-1 and ICAM-1 in certain domains and its principle ligand is the a4137

integrin on T and B lymphocytes which can bind to MAdCAM-1 or VCAM-1,

probably via the common a4 sub-unit (Leung et al., 1996). L-selectin is also a known

ligand for MAdCAM-1. The role of MAdCAM-1 in the periodontal disease immune

response is still not clear.

The migration of leukocytes from blood vessels into the tissues is called

extravasation. The whole process can be described in four different steps. The first

step involves weak adhesion of leukocytes to the vessel wall. Circulating leukocytes

are seen "rolling" along the area of endothelium which is involved in the

inflammatory response. These weak adhesions are mediated by the adhesion

molecules P-selectin and E-selectin which are found on endothelium cells. P-selectin

is expressed after a few minutes of exposure to TNF-a, and E-selectin after a few

hours. All selectins bind leukocytes after recognition of specific carbohydrate

epitopes on their surface; for example the Siayl-Lewisx moiety on specific

glycoproteins.

The second step of this migration process involves interactions between the

immunoglobulin related molecule ICAM-1 and LFA-1 and Mac-1. ICAM-1 is

induced by TNF-a on the surface of endothelial cells. LFA-1 and Mac-1 are always

found on the surface of phagocytic cells, however, they do not have a high affinity for

ICAM-1. IL-8 is a chemotactic factor for PMN inducing them to leave the

bloodstream and migrate into the surrounding tissue, in this case the gingival tissues.

IL-8 induces phagocytes to leave the blood vessels by greatly increasing the affinity of

39

LFA-1 and Mac-I for ICAM-1. This causes the phagocytes to bind tightly to

endothelium, and the rolling to arrest.

The third step involves the cell actually crossing the endothelial wall. This is due to

adhesive interactions involving LFA-1 and Mac-1 found on the leukocytes and also

PECAM or CD31, an immunoglobulin related molecule found both on the leukocyte

and also at the junctions of the endothelial cells. These interactions allow the

leukocyte to cross the vascular endothelial cell wall by a process called diapedesis.

The final step, is the migration of leukocytes through the tissues to the site of

infection. This occurs along a chemotactic gradient set up by chemotactic factors.

40

Table 1.1 Major Adhesion Molecules

NAME SYNONYM CD number)

LIGAND(S) RECEPTOR EXPRESSION

SELECTINS Initiate leukocyte to endothelium interaction L-Selectin LAM-1, Mel-14,

(CD62L) G1yCAM-1, MAdCAM-1, (CD34), Siayl Lewisx moiety

T-cell, Monocyte, Neutrophil

E-Selectin ELAM-1, (CD62E) Siayl Lewis' moiety Endothelium P-Selectin GMP-140, PADGEM,

(CD62E) PSGL-1, Siayl Lewis' moiety

Platelets, Endothelium

MUCIN LIKE VASCULAR ADDRESSINS Bind to L-selectin to initiate leukocyte-endothelial interactions CD34 L-Selectin Endothelium Gl CAM-l L-Selectin HEV Endothelium MAdCAM-1 (Immunoglobulin supergene member also)

L-Selectin/a4ß7/CS-1 Mucosal Lymphoid Tissue Venules

INTEGRINS Bind cell adhesion molecules and extra cellular matrix strongly ß1 (VLA)

a4(31 VLA-4 (CD49d/29) VCAM-1, Fibronectin, Peyer's Patches High Endothelial Venules

T-cell, B-cell

a5(3l VLA-5 (CD49e/29) Fibronectin T-cell, Endothelium, E ithelium, Platelets

ß2

aL 2 LFA-1, (CDIIa/18) ICAM-1, ICAM-2 Leukocytes

amß2 MAC-1, CR3, CD1lb/18)

ICAM-1, C3bi Fibronectin

Monocytes Neutro hils

I7

a4 p7 LPAM-1 (CD49d/-) MAdCAM-1, V CAM-1 T-cell, B-cell IMMUNOGLOBULIN GENE SUPER-FAMILY Various roles- Tar et or Integrins ICAM-1 (CD54) MAC-1, LFA-1 All cells ICAM-2 (CD 102) al 2 Endothelium PECAM-1 (CD31) PECAM-1 Endothelium VCAM-1 (CD106) a4 1 Endothelium LFA-2 (CD2) LFA-3 T-cell

41

1.4.8 Polymorphonuclear leukocytes (PMN)

PMN are granulocytes and along with basophils, mast cells and eosinphils are short

lived phagocytes which play an important role in the inflammatory immune response.

Under normal conditions PMN comprise between 50% and 70% of the circulating

leukocyte pool in man (Miller et al., 1984). The half-life of the PMN in the peripheral

circulation is brief, being 6-7 hours (Murphy, 1976).

The PMN plays a key role in the early stages of an inflammatory response and arrives

at the site of infection within a few hours. As previously discussed, when functioning

against an infectious agent, PMN leave the blood vessels and move across tissues

towards a chemotactic source. Adhesion molecules necessary for the movement of

PMN from vessels are expressed on endothelial cells and PMN (Rosales and Brown,

1993). The adhesion of PMN is controlled by upregulation of P-selectin, on

endothelial cells of post capillary venules by thrombin or histamine (Staunton et al.,

1990).

PMN respond to many different chemoattractants, including N-formyl peptides,

complement derived C5a-desarg, Leukotriene B4, IL-8, TNF and platelet activating

factor (Havarth, 1991). Using time-lapse cinematography or other visual methods for

following the tracks of cells, it has been shown that PMN can migrate in fairly straight

paths towards the source of a chemical gradient (Zigmond, 1974; Allan and

Wilkinson, 1978). Once PMN have been stimulated by chemoattractants they

undergo morphological changes from their unstimulated spherical morphology.

Within two or three minutes, the typical locomotor morphology changes to a ruffled

anterior lamellipodium, tapered cell body, with a tail (Smith et al., 1979; Shields and

Haston, 1985; Keller et al., 1983). As the cell moves forward, the most active

movements are seen at its head, and contraction waves may pass down through the

body of the cell to the tail (Haston and Shields, 1984). There is also evidence that

some receptors of the PMN may redistribute to the head of moving cells (Wilkinson,

1990). The mechanism for this is unknown. One theory suggests that a signal from a

ligand causes actin polymerisation. Thereafter, the newly formed cytoskeleton in the

42

pseudopod sweeps forward not only activating the receptor-ligand complex, but also

other membrane proteins (Wilkinson, 1990).

Once at the site of the chemotactic stimulus and prior to attachment of the PMN to the

invading organism, opsonisation takes place either in the presence or absence of

complement and antibody. The presence of opsonins, as mediated by PMN membrane

receptors, dramatically increases the rate of phagocytosis by PMN. Heat stable

opsonins include all 4 classes of Immunoglobulin G (IgG) (Miller et al., 1984) and the

heat labile opsonin, attributed to complement cleavage protein C3b (Stosse], 1974).

The products of C3 activation and IgG bind their respective PMN receptors, CRI and

CR3 for complement and FcyRII and FcyRIII for IgG. The only drawback of this

mechanism is when the host is dealing with a capsular organism. Capsular organisms

are able to evade this opsonisation process, and hence phagocytosis. This is due to the

masking of the antigenic portion of their cell wall by the capsule which avoids the

deposition of complement on their surface (Slots and Genco, 1984).

If antibody or complement are not available then alternative mechanisms of

opsonisation of the bacteria come into play. Lipopolysaccharide binding protein

(LBP) is an acute phase reactant which binds LPS, a cell wall component of Gram

negative bacteria themselves. Opsonisation is achieved in this way, as the binding of

LBP to LPS on bacteria enables ligation of bacteria to PMN through the CD14

receptor, thus enhancing phagocytosis (Wright et al., 1989).

Following opsonisation, phagocytosis can take place. Phagocytosis is the process

whereby the PMN takes up a particle. Adherence of the phagocytic cell to a pathogen

triggers a system of contractile actin-myosin filaments inside the phagocyte. This

enables the cell to "throw its arms" or pseudopodia around the target and enclose it in

its membrane. The membrane surrounding the pathogen fuses with membranes

surrounding lysosomes in the phagocytic cell. Lysosomes are small intracellular

packets of powerful degradative enzymes and toxic molecules, which are emptied

onto the engulfed pathogen with a destructive effect.

43

Once phagocytosed and engulfed by the PMN, there are two mechanisms for killing

the target organism; the oxygen dependent and the oxygen independent pathways

(Van Dyke, 1985). Oxygen dependent killing is activated when the PMN undergoes a

respiratory burst coincident with phagocytosis. Nicotinamide adenine dinucleotide

hydrogen phosphate (NADPH), a continuous supply of which is generated from

glucose via the hexose monosulphate shunt in the phagolysosome membrane, is

activated and reduces oxygen (02) to superoxide (02) (Weiss et al., 1982). Following

the generation of hydrogen peroxide (H202) and hydroxyl ions (OH), a second enzyme

system involving myeloperoxidase becomes activated. Myeloperoxidase is released in

large quantities, however, on its own has little effect. In combination with H202,

myeloperoxidase is capable of oxidising halides to reactive toxic compounds like

HOC13 and chloramines (Weiss, 1989).

OH is considered the most toxic of all the free radicals. This reaction in vivo is

limited by the availability of iron. Little free iron is actually available due to its

binding to transferrin and lactoferrin extracellularly and apoferrin intracellularly

(Sanchez et al., 1992). In the phagocytic vacuole the pH is reduced and some iron

may be lost from transferrin. Lactoferrin retains its ability to bind iron at low pH, thus

protecting the cell and limiting the production of OH (Halliwell and Gutteridge,

1985).

Oxygen independent killing is based on the PMN granule network. Antimicrobial

agents, which are independent of oxygen, are released into the phagolysosome during

phagocytosis (Thomas et al., 1988). These granule components, include

bactericidal/permeability-increasing protein, chymotrypsin-like cationic protein and

defensins (Weiss et al., 1982) and those outlined above. They are strongly

bactericidal and in most cases kill bacteria (Miyazaki et al., 1997). In addition to

killing bacteria, the proteolytic enzymes, including elastase and the

metalloproteinases, have the potential to damage extracellular tissue.

Many investigators have addressed the issue of whether the presence of PMN in the

sulcus indicates anything about the pathogenesis of periodontal disease. An

experimental study was carried out by Löe (1961). This investigated healthy gingival

44

sulci in a group of dogs that were sealed at the margin. The results of the study

indicated that neutrophils and epithelial cells clustered near the margin and had filled

the sulci in 12-48 hours. This study demonstrated that the increase in trapped material

was due to the continuing phenomenon of epithelial desquarnation and migration of

neutrophils that occurs in normal situations. It is generally accepted that PMN

migrate into clinically healthy sulci in small numbers. One study which supports this

theory, was carried out by Yamasaki et al. (1979). This report showed that PMN were

routinely seen within the intercellular spaces of the junctional epithelium of germ free

rats, at all levels of attachment. This investigation was carried out on germ free rats

and therefore, indicted that the presence of PMN in the sulcus cannot be exclusively

explained on the basis of the presence of micro-organisms. It may represent a

physiological phenomenon specific to this tissue type.

There is a statistically significant increase in PMN in the junctional epithelium of

humans in experimental gingivitis trials after 2-4 days, compared to clinically healthy

gingiva (Payne et al., 1975). In this study, as gingivitis developed, the number of

leukocytes also increased, until oral hygiene measures were reinstated. Cell counts

showed that 95-100% of the cells collected were PMN. Very few leukocytes were

found in the crevices of healthy gingiva.

Animals have been used over the last few decades for immunological studies to

provide evidence for the function and importance of PMN in the periodontal

environment. Garant (1976) showed that in monoinfected rats, the PMN response to

micro-organisms resulted in formation of a neutrophil wall or barrier between the

plaque and host tissue. It has been suggested that this barrier may assist in preventing

invasion of bacterial products into the tissues.

Further indications that PMN play a role in the inflammatory immune response, come

from patients that suffer from periodontal disease, as well as other clinical

manifestations, particularly deficient PMN function. Neutropenic states have been

associated with rapidly progressive forms of periodontal disease (Miller et al., 1984).

In cyclic neutropenia, every 14-21 days, PMN disappear from the circulation and

reappear after approximately 5 days. In health, PMN comprise 50% to 70% of the

45

circulating leukocyte pool, whereas in cyclic neutropenia they never comprise more

than 50%. Recurrent ulcerations of the oral cavity are common, particularly on the

mucosa, the lips and the throat, whenever the number of PMN fall. Radiographs have

been taken in some cases, these have indicated involvement of the periodontal tissues,

showing bone loss around the primary and secondary teeth (Miller et al., 1984).

A high percentage of family members, with a background of localised aggressive

periodontitis, have been reported to show abnormal chemotaxis (Van Dyke et al.,

1985). When 22 families with localised aggressive periodontitis were analysed, it was

shown that in 19 of these families, the localised aggressive periodontitis patient and

their family members all had neutrophil chemotaxis disorders. In all of the 22

families the chemotactic defect was shown in 50% of the siblings, indicating a

dominant mode of inheritance. The cause of the defect remains controversial, being

due to either intrinsic defects in the PMN itself or extrinsic factors in the sera is

unknown (Agarwal et al., 1996). If the neutrophil chemotactic defect is due to a

major gene locus, heterogeneity exists as there is a significant number of patients and

families that show no evidence of the defect (Kinane et al., 1989a; Page and Beck,

1997)

1.4.9 The adaptive immune system

It is unclear how many infectious agents encountered by an individual, are dealt with

and eliminated by our innate immune system. For the adaptive arm of the immune

system to be triggered, infection has to elude the innate defence mechanisms and

exceed the threshold dose of antigen required to initiate the adaptive immune

response. The adaptive immune response can be divided into two distinct entities; the

humoral immune response and the cell mediated immune response. Both of these

arms of the immune system are adaptive, both involve the single most important

principle in adaptive immunity, which is clonal selection of lymphocytes, and both

lead to a long lasting protection. The divergent lifestyles of different pathogens lead

to the requirement of different mechanisms for their recognition and elimination.

46

1.4.10 The humoral immune response

The antibody was the first product of the specific immune response to be identified. It

was found in plasma of blood, and at that point body fluids were called humors, so the

antibody immune response became known as the humoral immune response (Janeway

and Travers, 1994). B lymphocytes are small lymphocytes that give rise to antibody-

producing cells. They are called B lymphocytes or cells because they emanate in the

bursa of Fabricius. This is a central lymphoid organ for B cell development, and

found only in birds and the bone marrow in mammals. B cells are stimulated by

antigen to produce and secrete antibodies, however, at the point when antibodies are

produced, the B cells have differentiated into plasma cells. The regulation of

immunoglobulin synthesis can be divided into two different processes; the control of

the maturation of stem cells into B cells, and the regulation of the terminal maturation

of B cells into immunoglobulin secreting plasma cells, a process which is under the

control of a complex network of regulatory T cells (Waldmann, 1983). Plasma cells

differ from B cells, as they have cytoplasm characteristic of an active secretary cell. It

is dominated by layers of rough endoplasmic reticulum and a prominent golgi

apparatus. The interaction between an antibody and an antigen comes about solely on

the basis of specific antigen recognition (Janeway and Travers, 1994).

Antibodies of all specificity's belong to a family of plasma proteins called

immunoglobulins (Igs). Five distinct classes of immunoglobulin or isotypes exist;

IgM, IgD, IgE, IgA and IgG (McGhee and Kiyono, 1999). These different classes of

immunoglobulin can be distinguished biochemically and functionally. Each antibody

is a `Y' shaped molecule whose arms form two identical antigen binding sites, that are

highly variable between one molecule and another. An example can be seen in figure

1.1. The stem of the `Y' shaped molecule has limited diversity and is known as the

constant region. The structure of the `Y' shaped molecule has been determined using

x-ray crystallography.

47

Figure 1.1 Structure of Immunoglobulin IgGI

Regions that bind

antigen

F

Cell receptor-binding sites

Complement binding domains

LIVariable regions

Fc

Adapted from the Department of Immunology, University of Glasgow web site.

Constant regions

The immunoglobulin molecule is made up of 3 equally sized globular domains and

these are held together by a hinge region, that is a stretch of polypeptide chain. Each

arm of the `Y' is formed by the association of a light polypeptide chain of

approximately 25 kDa, with the amino terminal half of a heavy polypeptide chain of

approximately 50 kDa. There are 2 types of light chain that exist. These have been

termed lambda (2) and kappa (K). No functional difference has been found between

antibodies that possess k chains or x chains, however, the ratio with which they are

used does vary between species. There are 5 types of heavy chain or isotypes that

exist as previously mentioned. The distinct functional properties of the 5 isotypes are

conferred by the carboxy-terminal half of the heavy chain of the molecule. From

protein sequencing of the heavy and light chains of numerous different antibody

molecules, it has been revealed that there is substantial sequence variability at the N-

terminal ends of the chains. These regions of the heavy and light chains have thus

been named the variable regions. Conversely, the C-terminal ends of both the light

and heavy chains are considerably similar and have therefore, been termed the

constant regions. Variable regions of the molecule correspond to the parts of the

antibody that recognise and are specific for antigens. Antigens are found to bind and

lie in the cleft formed by the interface of the heavy and light chain variable domains,

specific for that antigen (Janeway and Travers, 1994).

Regions of antigens recognised by antibodies are called antigenic determinants or

epitopes. When an epitope is recognised by a specific antigen binding portion of an

immunoglobulin an interaction is made. A number of environmental factors can

disrupt interactions, examples of which are extremes of pH, high salt concentrations

and detergents. Interactions are held together by non-covalent forces the sum total

strength of which is called the avidity. Hence, the above mentioned environmental

factors can disrupt the interaction and decrease the avidity.

As previously mentioned, differentiation from aB cell into an immunoglobulin

secreting plasma cell, requires the co-operation of other cell types. The current theory

indicates the requirement of B cells, T cells and macrophages, implying that aB cell

requires more than one signal for activation, as well as the essential recognition signal

of an antigen. The first contact between antigen and the B cell evokes a primary

49

antibody response. This has the characteristically long lag period between stimulus

and detection of specific antibodies, an exponential increase in antibody before it

reaches equilibrium after some days, followed by a steady decline (Virella, 1990). In

the primary antibody response, the first antibody class to be synthesised is usually

IgM. The activation of these small naive resting B cells is a multistep process. The B

cell leaves the bone marrow expressing on its surface both IgM and IgD, which are

receptors specific to that B cell. They also express major histocompatability complex

(MHC) II molecules and CD45 (Snow and Noelle, 1993). After leaving the bone

marrow the naive B cells circulate around the body, between the blood and the

lymphatics, before they either converge with their specific antigen or die. Most

peripheral B cells have a life span of less that two weeks (Freitas et al., 1986;

Udhayakumar et al., 1988).

During the humoral immune response, B cells are activated through the recognition

and binding of their immunoglobulin receptor with that of an antigen. This activation

leads to a differentiation process which results in the production of either plasma cells

or long lived memory cells (Berek, 1992). With some bacterial antigens, for example

surface polysaccharides, the process of plasma cell differentiation is independent of T

cell help. The activation of B cells by protein antigens however, induces a more

complex set of reactions (Berek, 1992).

For activation to occur, aB cell must interact with aT cell also specific for the

antigen, as well as with the antigen itself (Vitetta et al., 1989; Noelle and Snow,

1990). This mechanism may have arisen because the regulation of B cells is not as

stringent as for T cells. Surface immunoglobulin specificity is generated through

random gene rearrangements. This makes the existence of B cells expressing surface

immunoglobulin reactive with self antigens possible. T cells, however, have to

undergo extensive selection in the thymus, to try and ensure all the self-reactive T

cells are removed. Therefore, to activate aB cell, an antigen also has to be processed

and recognised by aT cell. Consequently, the lack of self-reactive T cells can directly

limit B cell responses to self antigens (Myers, 1991).

50

The B cell receptor, which as previously discussed is the immunoglobulin molecule,

recognises epitopes on the native antigen and has the same specificity as the antibody

that is eventually secreted by the cell (Vitetta et al., 1989). However, before most

antigens can be presented and recognised by T cells, they have to be processed (Allen,

1987). Protein antigens can be processed by B cells and other antigen presenting

cells. T and B cells, both with specific receptors for the antigen can collaborate by

cross-linking of the immunoglobulin and T cell receptor antigen receptor complexes

(Reth, 1992; Weiss and Littman, 1994). The interaction of T cells with antigen-

activated B cells and antigen presenting cells initiates a cascade of reactions which

lead to the maturation of the immune response and development of memory cells

(Berek, 1992).

There seem to be two discrete developmental pathways following primary antigenic

exposure of B cells (Maclennan and Gray, 1986). The first pathway involves a rapid

clonal expansion and plasma-cell differentiation of the cells in the T-cell rich areas of

the secondary lymphoid organs. These clonally expanded plasma cells thus, carry out

the immediate effector function, the clearing of target antigens (McHeyzer-Williams

and Ahmed, 1999). The second pathway involves extensive clonal expansion within

the specialised micro environment of the germinal centre (GC). GCs are sites of

intense B-cell proliferation. A GC reaction is observed only where T helper cells are

involved (Kroese et al., 1990).

Therefore, B cells first bind antigen specific to their immunoglobulin receptors, and

are activated by helper T cells in the T cell areas of lymphoid tissues. This area is

known as the periarteriolar lymphoid sheath, or PALS (Kelsoe, 1996). At this point B

cells or centrocytes, as they are known, proliferate in the PALS and then either

develop locally into foci of antibody secreting plasmacytes, or migrate to adjacent

lymphoid follicles to initiate the GC reaction (Kelsoe, 1996). The majority of early

primary antibody is produced by these plasmacytic foci. They are able to secrete

unmutated IgM or IgG over the following 10-12 days, thus providing an early source

of circulating antibodies. These cells eventually undergo apoptosis and disappear

(Jacob et al., 1991 a; Jacob et al., 1991 b; Jacob and Kelsoe, 1992; Smith et al., 1996).

51

Other activated B cells migrate into the primary follicles and form GCs. Primary

follicles contain resting B cells clustered around a dense network of follicular

dendritic cells (FDC), a specialised cell type in this area. The activated B cells

accumulate amongst the FDC which bear antigen-antibody and antigen-complement

complexes on their surface (Kelsoe, 1996). The activated B cells rapidly proliferate

and acquire new phenotypic characteristics (Kelsoe, 1996). Over the period of the

following 14 days, these B cells give rise to a GC developing into at least two regions:

a dark and a light zone (Maclennan, 1994).

Hypermutation of the variable region of the immunoglobulin molecules and selection

of high affinity variants occurs in the GC. This is of key importance to the maturation

of the immune response in this environment (Berek and Ziegner, 1993). Somatic

hypermutation affects all the rearranged variable-regions in aB cell, allowing the

generation of variability (Janeway and Travers, 1994). This is due to a molecular

mechanism that introduces point mutations into V regions. The recruitment of aB

cell into the maturation process is dependent on the affinity of its receptor for antigen

(Berek and Ziegner, 1993). One view is that the initial selection of these cells is based

upon their continuous ability to bind antigen from the surface of the FDC, as an

antigen-antibody complex (MacLennan et al., 1990; MacLennan et al., 1992; Liu et

al., 1992). Efficient affinity selection is only thought to occur when cells that have a

higher affinity for antigen than the antibody within the FDC immune complex are

stimulated (MacLennan and Gray, 1986). This makes sense considering the memory

B cell response is of higher affinity than the primary response (Gray, 1993).

Following this positive selection in the GC reaction, affinity matured B cells can

either exit the GC as memory cells or sometimes re-enter the dark zone of the GC for

a further round of mutation and selection (Berek and Ziegner, 1993)

It is generally accepted that the cells produced in the GC are memory cells, which then

migrate into the circulation to be stimulated on a second encounter with the same

antigen (Hanna et al., 1968; Jacobson and Thorbecke, 1968; Wakefield and

Thorbecke, 1968a; Wakefield and Thorbecke, 1968b). Studies have shown that cells

produced from the GC are not likely to be involved in antibody production, since there

52

is no correlation between the production of cells in the GC and serum levels of

antibody (Buerki et al., 1974).

At the beginning of the humoral immune response, the predominant antibody isotype

is IgM (Snow and Noelle, 1993). Later on in the immune response, the B cells are

able to undergo isotype switching and other immunoglobulin isotypes are seen. This

phenomenon is important to the humoral immune response, as each different isotype

possess unique capabilities that improve the effector capabilities of the organism

against pathogens (Snow and Noelle, 1993). Isotype switching occurs during the GC

reaction (Kraal et al., 1982; Butcher et al., 1982; Ape] and Berek, 1991).

Antibodies have different functions, all of which are centred around protecting the

host from extracellular pathogens and their products. Antibodies have a number of

ways of providing protection.

Neutralisation is a process whereby antibody binds to the surface of a pathogen itself,

or to a toxin or product produced by the pathogen. Intracellular bacteria and viruses

survive and reproduce inside cells. However, in order to increase in number they need

to be transferred from cell to cell and to accomplish this, these bacteria and viruses

bind to cell surface molecules of other host cells. Specific antibodies are able to bind

to the bacteria or virus particles and neutralise them, preventing cell binding and

spread. Many bacteria cause devastating effects by secreting molecules called

bacterial toxins. The toxins bind to a molecule that acts as a receptor, on the surface

of a host target cell. Antibodies specific for the receptor portion of the toxin, can

block binding to a host cell and protect it from toxic attack. Although neutralisation

prevents adherence of pathogens and their products to host cell surfaces, it does not

remove the pathogens from the host altogether. Furthermore, many pathogens are not

neutralised by antibodies. They have developed strategies whereby they can avoid this

defence mechanism.

Other bacteria and pathogens reproduce and survive extracellularly. These pathogens,

along with those that cannot be neutralised, need to be removed. Antibodies protect

53

against these by facilitating their uptake into phagocytic cells, such as PMN, by a

process called opsonisation. There are two ways in which this can occur.

Opsonisation can occur when an antibody binds to a micro-organism by its Fab

portion. The Fc portion of the antibody is recognised and bound by a receptor found

on PMN. The micro-organism then becomes engulfed by the phagocyte as receptors

for Fc and the Fc regions on the antibodies continue to combine, leading to final

engulfment and destruction of the micro-organism. The second way in which

opsonisation can occur is by activation of the complement cascade and binding of

phagocytes to complement components found on the surface of the micro-organism.

1.4.10.1 IgM

The first antibodies to be produced in an immune response are IgM. Antibodies of

this isotype are produced before B cells have undergone somatic hypermutation and

affinity maturation, and hence tend to be of low affinity. However, IgM molecules are

usually found as pentamers with 5 immunoglobulin molecules and 10 antigen binding

sites. This compensates for the low affinity binding of these antibodies, as it allows

simultaneous binding of antibody to repetitive sites, which are often found to be

expressed by bacterial cell wall polysaccharides. This binding to multimeric antigens

can confer high avidity.

IgM is a large molecule and the pentamer is formed in association with aJ chain and

for this reason it does not pass through the placenta. The monomers of IgM are cross-

linked to each other by disulphide bonds and also to the J chain. Due to the large size

of the IgM molecule, it is usually confined to the blood, however, IgM can be found in

tissues at sites of infection that are undergoing an inflammatory response. Elevated

levels of IgM usually indicate either recent infection or exposure to antigens.

One of the most important functions of IgM is that it is very potent at activating the

complement cascade. Activation of these plasma proteins is extremely important,

especially at the beginning of an infection as it helps recruit and activate phagocytes

and can directly destroy pathogens (Janeway and Travers, 1994). IgM antibodies are

54

not particularly versatile, are poor toxin-neutralising antibodies and are not efficient in

the neutralisation of viruses.

1.4.10.2 IgA

IgA is the major mucosal immune system isotype made by plasma cells and has a half-

life of 5.5 days. The primary site of IgA synthesis is at the epithelial surfaces of the

body and its main role is thought to be to protect epithelial cells from infectious

agents. IgA achieves this by preventing attachment of bacteria and toxins to epithelial

cells, and is involved in the neutralisation of harmful toxins. IgA provides the first

line of defence for these surfaces. IgA is a poor opsonin and does not contain

receptors for complement, and thus is not a complement-activating or complement-

fixing immunoglobulin. This is not surprising as IgA is usually found in areas which

are lacking in accessory cells.

The lamina propria (connective tissue), that is found beneath the basement membrane

of many surface epithelia, hosts a large number of IgA secreting plasma cells

(Heremans, 1974). Active transport or transcytosis of IgA occurs from the lamina

propria, where it is produced, to the external surface of the epithelium. Secretary IgA,

or IgA that undergoes transport is predominantly of the polymeric form (Heremans,

1974). Polymeric IgA is usually found as a dimer with a molecular weight of 400

kDa. Two IgA molecules are held together by disulphide bonds and are both also

bound to aJ chain, found also in pentameric IgM. The J chain, produced by plasma

cells has been implicated as a component for the polymerisation of IgA and IgM

(Koshland, 1975).

There are 2 subclasses of IgA; IgAl and IgA2. IgAl and IgA2 are distributed

differently in the body fluids; 80% or more of serum IgA belongs to the IgAl

subclass, whereas 50-74% of the IgA found in external secretions is of this subclass

(Vaerman et al., 1968; Grey et al., 1968; Delacroix et al., 1982). IgA2 is found in a

higher percentage in external secretions. The reason for this is unknown, but it could

be due to the preferential transport of IgA2, or that increased numbers of IgA2

secreting cells are found at secretary surfaces.

55

IgA antibodies are secreted in breast milk and transferred to the gut of new born

infants. Their specialised defence of mucosal surfaces provide protection of the new

born until they can provide their own.

1.4.10.3 IgG

IgG molecules are small and able to diffuse out of the blood and into the tissues. IgG

molecules are always monomeric and are the principal isotype in the blood and

extracellular fluids.

There are 4 different subclasses of IgG; IgGI, IgG2, IgG3 and IgG4. Except for IgG3

which has a rapid turnover, the half-life of IgG is the longest of all the isotypes at

about 23 days. Functional differences can be noted amongst the different isotypes.

IgG2 subclass is more frequently found to be produced in response to polysaccharide

antigens, IgGI and IgG3 are involved in defence against protein and viral antigens,

and IgG4 is mainly associated with responses that can lead to allergic type reactions.

IgGI, IgG2 and particularly IgG3 play an important role in the activation of the

classical pathway of the complement cascade. IgGI and IgG3 are the only subclasses

that are found to bind to macrophages and other phagocytes, and hence play an

important role in the process of opsonisation. IgG are high affinity antibodies, as they

are produced by B cells that have undergone affinity maturation and isotype switching

in the GC reaction. They also have the ability to cause neutralisation of toxins and

agglutination.

IgG molecules play an important role in antibody-dependent cell mediated

cytotoxicity (ADCC). This process involves binding of the Fab portion of the

antibody with the target cell, and the Fc portion binds with specific receptors found on

natural killer (NK) cells. The target cell is then killed, not by opsonisation but by

specific release of harmful molecules produced by NK cells.

56

IgG is the only antibody that is transported across the placenta. It is actively

transported directly into the bloodstream of the foetus by a specific IgG transport

protein. The small size and high affinity nature of the IgG molecule makes it an ideal

isotype for this function. IgG2 is the only isotype of IgG that does not cross the

placenta (Janeway and Travers, 1994).

1.4.10.4 IgD

IgD has a half-life of 2.8 days and is found on the surface of B cells at different stages

of their maturation. The function of IgD remains elusive.

1.4.10.5 IgE

IgE has a half-life of 2 days, which is the shortest of all the immunoglobulins. It is

also found at the lowest concentration in the serum. IgE cannot cause agglutination or

neutralisation, and has very specific functions in the immune surveillance of the body.

IgE antibodies are thought to play an important role in parasitic infections and

hypersensitivity reactions. IgE antibodies are able to bind directly to mast cells and

basophils by a receptor. The receptor is found on the surface of these cells and is

specific for the Fc portion of the IgE molecule. Cross-linking of IgE to an antigen via

its Fab portion, and a basophil or mast cell via its Fc portion, leads to activation of

these leukocytes. Activated mast cells and basophils release the contents of their

granules which include; histamine, heparin and leukotrienes and these can trigger

hypersensitivity reactions (Benjamini et al., 1996). The Fc receptors found on mast

cells and basophils, which are designated FcEI receptors, bind the Fc portion of IgE

with very high affinity.

Degranulation occurs within seconds of cross-linkage. First the vasoactive amines,

histamine and serotonin are released which leads to the initiation of an inflammatory

immune response. There is increased vasodilation and vascular permeability, which

quickly leads to the accumulation of immune cells and exudative fluids at the site of

infection. Serotonin and histamine are very short lived, but the inflammatory response

57

is sustained by the subsequent release of other mediators by the mast cells and

basophils such as leukotrienes and other metabolites of archidonic acid (Janeway and

Travers, 1994).

1.4.11 The role of the adaptive humoral immune response in periodontal disease.

Two distinct humoral effector systems that contribute to the defence in the oral cavity

have been postulated by Brandtzaeg (1972,1973). Salivary IgA has been suggested to

be the first line of defence, perhaps playing a role which involves trapping of the

antigen in a mucin layer with subsequent disposal of the antigen (Brandtzaeg, 1972).

The second involves the local production of IgG which is protective in nature and

found in inflamed gingival and periodontal tissues (Brandtzaeg, 1973). These early

ideas are still postulated in publications on this topic, and are often referred to as the

local systemic immune responses. The importance of these different fluid types and

what they actually indicate is still to be confirmed. In addition, it is not fully

understood whether the local levels, such as that of saliva and GCF, are reflections of

the serum antibody levels or whether production of the antibodies found in such fluids

is distinct and unique.

1.4.12 The local antibody response

In terms of the local immune response, saliva is known to contribute to maintaining

oral health, and secretary IgA is found in large quantities in saliva. It has been

suggested that an individual with high levels of IgA in their saliva, which is directed

against micro-organisms such as A. actinomycetemcomitans, P. gingivalis and

Streptococcus mutans, are more likely to be protected against gingivitis (Shenk et al.,

1993). Another study carried out by Sandholm et al. (1987), found elevated levels of

salivary IgG in patients with chronic and aggressive periodontitis against A.

actinomycetemcomitans.

Various studies have reported an increase in antibody levels in the GCF (Reinhardt et

al., 1989; Wilton et al., 1993) and in the gingival tissues (Smith et al., 1985). Further

studies have detected specific antibody activity in the GCF of patients with

58

periodontal disease (Ebersole and Cappelli, 1994; Ebersole et al., 1985a; Ebersole et

al., 1985b; Smith et cal., 1985). Through clinical observation of diseased tissues taken

from periodontitis patients, it is known that they are populated by a large number of B

lymphocytes and plasma cells, thus suggesting local antibody production (Ranney,

1991; Sims et al., 1991). These cells are seen to differentiate and proliferate in this

area, and in susceptible subjects this area is also populated by an increased number of

TH2 cells that have selectively homed to the gingiva to aid in B-cell expansion

(Seymour et al., 1993). More evidence for local antibody production comes from

reports indicating the presence of activated complement components in GCF from

both the classical and alternate pathways (Niekrash and Patters, 1985; Schenkein and

Genco, 1977). This further indicates the possibility of local IgG synthesis in the

tissues.

Many of the studies observing the presence and function of antibody in GCF have

been particularly involved in looking at antibodies directed against A.

actin onrycetemcoin itans and P. gingivalis. A study carried out by Ebersole and

Cappelli (1994) investigated antibodies directed against A. actinomycetemcomitans

and the presence of different subclasses of antibody in GCF of patients with

periodontal disease. They compared the findings with those from subjects in health.

The frequency and distribution of antibody in the GCF indicated a potential role for

local antibody directed against A. actinomycetemcomitans in relation to colonisation

and clinical presentation. The results also suggested a relationship between the

distribution of infection and the levels of antibody directed against A.

actinomycetemcomitans in GCF.

The results gleaned from studies observing antibodies directed against P. gingivalis

give varying reports. For example, Baranowska et al. (1989) reported no difference in

antibody levels against P. gingivalis in the GCF of periodontitis patients compared

with health. However, Suzuki et al. (1984), indicated there was an increase in the

local production of antibodies directed against this micro-organism in patients with

chronic periodontitis as compared with patients with localised aggressive

periodontitis.

59

Although the results from a lot of the work carried out in this area indicate that there

is a local immune response due to the micro-organism in question providing antigenic

stimulation at a local site, there is also a systemic serum antibody response to the local

infection. A relationship between the local and systemic antibody levels and

specificity is thought to exist. Theories have been put forward trying to link the local

and the systemic immune responses. Three main theories that have been suggested by

Kinane et al. (1999): (i) local antibody is produced due to continuous antigenic

challenge, however, this antibody leaves the tissues via the lymphatics and enters the

circulation becoming systemic; (ii) a systemic response is induced due to a breakdown

in the tissue integrity of the gingivae, allowing bacteria and antigenic components to

leave, migrate to the lymph nodes in the circulation, and induce a systemic response;

(iii) local antigen presenting cells pick up and process antigens in the tissue before

leaving and homing to systemic lymphoid tissues, where they induce a systemic

antibody response.

Studies have reported GCF levels of antibody to be greater than levels found in the

serum (Ebersole et al., 1985a; Kinane et al., 1993; Tew et al., 1985a). As well as

being a product of the local immune response, it has also been suggested that serum

antibody specificity may reflect a local gingival response. This has been suggested

following studies observing the response to A. actin onrycetemcoin itans in periodontitis

patients (Ebersole, 1990; Ebersole et al., 1992).

IgG, IgM and IgA classes of antibodies have been described in GCF (Brill and

Brönnestam, 1960; Brandtzaeg, 1965) and in the gingival tissues (Brandtzaeg and

Kraus, 1965; Thonard et al., 1966; Genco et al., 1974). Although these and serum

antibodies, against individual plaque and Gram negative bacteria have been shown to

be present in humans with varying degrees of periodontal disease, their specific

antigenic target remains on the whole elusive. Different subclasses of antibody,

particularly IgG, carry out various functions, and in general are directed against

different antigenic molecules. Therefore, a number of studies have been carried out to

look at the different subclasses of antibody, with the aim of providing further

information on the nature and target of the immune response.

60

A study carried out by Ebersole and Cappelli (1994) investigated the different

subclasses of IgG directed against A. actinomycetemcomitans in the GCF of patients

with periodontal disease. This study showed that IgG3 and IgG4 subclasses were

particularly increased as compared with the levels seen in the serum. The patterns of

subclass distribution were quite distinct. In GCF samples, where the antibody levels

were below those found in the serum, IgG2 levels were found to be high. In GCF

samples that had antibody levels in the same range as those found in the serum, IgGI

was strikingly increased. In GCF samples that had elevated levels of antibody

compared with serum, elevated levels of IgG3 and IgG4 were found overall. Results

showed that IgG3, IgG4, IgGI and IgG2 antibodies were elevated by 58%, 35%, 25%

and 25% respectively. The correlation between elevated levels of IgG4 antibody to A.

actinomycetemcomitans and the presence of the micro-organism in the plaque was

positive.

Antibodies in the oral cavity have been detected against non-oral bacteria (Berglund,

1971; Mallison et al., 1989). Some of the antibodies may be cross-reacting antibodies

induced to another bacteria in a different part of the body. However, even if this is so

for a percentage of the antibody response, it is generally accepted that there is a

specific humoral immune response, with specific antibody produced, that plays a role

in protection against periodontal disease.

1.4.13 The systemic antibody response

The majority of aggressive and chronic periodontitis patients mount a systemic

humoral immune response to antigens of the infecting bacteria (Ishikawa et al., 1997).

As discussed previously there are several theories on how a systemic response is

induced.

One of the first experiments on this topic was carried out by Genco et al. (1980a).

This study indicated that there were antibodies directed against sonicate preparations

of A. actinomycetemcomitans in the sera of patients with localised aggressive

periodontitis, but not in healthy subjects. Many studies that have been carried out

61

since these early experiments have tended to target the Gram negative anaerobes A.

actino1nycete1ncontitans and P. gingival s because of their marked association with

periodontal disease (Socransky and Haffajee, 1990). Systemic antibodies to a large

number of Gram negative bacteria have been investigated. F. nucleation appears to be

a regular resident in plaque. Although serum antibodies directed against this bacteria

seem to be detectable in patients with periodontitis, the titres are not elevated in all

periodontitis patients (Farida et al., 1986a; Suzuki et al., 1984; Tolo et al., 1981).

Elevated levels of serum antibody have been detected in selected periodontitis patients

against E. corrodens (Ebersole et at., 1987; Ebersole and Holt, 1988) and elevated

IgG antibody levels against C. rectus have been shown in all periodontitis patients

compared with health (Nisengard et al., 1980).

Most of the studies attempt to assess the antibody responses to A.

actinomycetemcomitans and P. gingivalis. A strong correlation has been shown

between increased levels of antibody to A. actinonrycetemcomitans and localised

aggressive periodontitis. It is now accepted that this micro-organism is important in

the causation of aggressive periodontitis. This is because of its high prevalence in the

disease and the elevated antibody levels in these patients against this bacteria

(Ebersole et al., 1980; Ebersole et al., 1989; Genco et al., 1980a; Farida et al., 1986;

Genco et al., 1980b; Tew et al., 1985a). Although in many diseases there appears to

be considerable variation in responses seen between different populations, global

results agree that there are elevated levels of serum antibodies to A.

actinomyceteincomitans in patients with aggressive periodontal disease, particularly

the localised form. The results of these different geographical studies seem to

suggest, that the predominant serotype of A. actinomycetemcomitans varies

(Gunsolley et al., 1988; Gmür and Baehni, 1997; celenligil and Ebersole, 1998).

Elevated serum antibody titres to P. gingivalis have also been identified in patients

with aggressive periodontal disease. A large study carried out by Ebersole indicated

that of the 62 generalised aggressive periodontitis patients observed, 37% of these had

elevated antibody titres to P. gingivalis, compared with only 25% in control subjects

(Ebersole et al., 1982a). Similar results were obtained from a study carried out by

62

Mooney and Kinane (1994) who showed that 25% of the generalised aggressive

periodontitis patients tested had elevated antibody levels.

In terms of chronic periodontal disease, A. actinomycetemcomitans does not appear to

be uniformly linked. Many studies have not found any elevated levels of serum

antibodies to A. actinoniycetenacoinitans in patients with chronic periodontitis

compared with healthy controls (Doty et al., 1982; Gmür et al., 1986; Listgarten et al.,

1981). Results from investigations by Ebersole et al. (1991a; 1994), have suggested

that increased titres of serum antibodies are seen, in some but not all, of patients with

chronic periodontitis. It is thought there is a group of individuals that are more

susceptible to infection by this bacteria than others, and that in this àt risk' group, A.

actinomycetemconaitans is capable of initiating disease.

Although serum antibody titres specific for A. actinolnycetemcomitans may not be

elevated in patients with chronic periodontitis, many studies have reported increased

antibody levels in these patients against P. gingivalis (Ebersole et al., 1982a;

Mansheim et al., 1980; Socransky and Haffajee, 1994). In addition there have also

been reports indicating a correlation between levels of IgG antibody to P. gingivalis

and periodontal destruction (Gmür et al., 1986), disease severity (Naito et al., 1987),

and alveolar bone loss in an elderly population (Wheeler et al., 1994).

Many studies have tried to evaluate the different subclasses of IgG in the serum of

periodontitis patients. This information may give an indication of the target of the

antibody response due to the functional differences noted amongst the different

subclasses. A number of studies by Ebersole et al. (1985c; 1989) investigated the IgG

subclass distribution in the serum directed against A. actinonrycetemcomitans and P.

gingivalis in patients with localised aggressive, generalised aggressive, chronic

periodontitis and also in healthy subjects. The results showed elevated levels of IgG3

and IgGI to A. actinonrycetemcomitans in patients with localised aggressive

periodontitis and elevations in IgG3, IgGI and IgG4 in patients with generalised

aggressive periodontal disease. Studies have also been carried out to investigate

levels of IgG2, a subclass that is found frequently in response to polysaccharide

antigens (Kinane et al., 1999a). Results from these studies indicated that aggressive

63

periodontitis patients seem to have elevated levels of serum IgG2, directed against

LPS of A. actinomycetemcomitans, and that these antibodies are opsonic for this

micro-organism (Gmür and Baehni, 1997; Wilson et al., 1995; Wilson and Hamilton,

1995).

Elevated levels of IgG2 antibodies have also been found in the serum of chronic

periodontitis patients against the LPS of P. gingivalis (Schenk and Michaelsen, 1987).

Moderate amounts of IgGl, IgG3 and IgG4 were also found in the serum of these

patients. Other studies have shown there to be elevated levels of IgG2, IgGI and IgG4

to P. gingivalis in patients with generalised aggressive and chronic periodontitis

(Ebersole et al., 1985c; Farida et al., 1986).

Evidence exists which may explain why high levels of IgG2 against P. gingivalis and

A. actinomycetemcoinitans are seen in patients with periodontal disease. Both of these

micro-organisms are Gram negative bacteria containing both LPS and a capsular

polysaccharide. As discussed in section 1.6.4 LPS is an immunogenic molecule and

as IgG2 is predominantly induced in response to polysaccharide antigens it is expected

that there would be elevated levels of IgG2 in patients mounting a response. In

respect to the distribution of the other subclasses of IgG, there appears to be enormous

variation between different populations and between the different forms of the

disease.

1.4.14 The cell mediated immune response

The principal components of the cell mediated arm of the immune system are T cells.

These lymphocytes are capable of recognising antigens, that are processed and

presented by antigen presenting cells (APC), in the context of the MHC (Babbitt et al.,

1985; Buus et al., 1987). The MHC is a set of polymorphic genes encoding class I

and class II cell-surface glycoproteins whose function is to present antigenic peptides

to CDS+ and CD4+ T cells respectively (Accolla et al., 1995).

Effector T cells fall into 3 functional classes that detect antigens derived from

different types of pathogens, presented by the two different classes of MHC molecule.

64

Antigens derived from pathogens that multiply in the cytosol are carried to the cell

surface by MHC class I molecules and presented to CD8+ cytotoxic T cells.

Histologically different cell types may be differentiated by cell-surface molecules.

These can be identified by a specific group of monoclonal antibodies. These

molecules are known as cluster of differentiation antigens and are designated CD.

Cells infected with viruses or bacteria that reside in the cytosol are eliminated by

CD8+ cytotoxic T cells. The function of these cells is to kill infected cells. Naive

CD8+ T cells can only differentiate into cytotoxic cells. They require more co-

stimulatory activity to activate them than do CD4+ T cells. This is possibly because

they are so destructive. Cytotoxic T cells principally act by, release of secretary

granules onto the surface of a target cell on recognition of an antigen (Janeway and

Travers, 1994) and the production of IFN-y.

Antigens derived from extracellular bacteria and toxins are processed within the APC.

They are carried to the surface by MHC class 11 molecules and presented to CD4+ T

cells. There are two distinct subsets of CD4+ T cells; THI and TH2. The division of

CD4+ T cells into subsets is based upon cytokine production (Mossman and Coffman,

1989). TH I cells secrete both IL-2 and IFN-y when activated by certain types of T-

dependent antigens so they can enhance cell-mediated responses. The TH2 subset

produces IL-4, IL-5, IL-10 and IL-13, and thereby promote the humoral immune

response. THO cells have also been identified and these are thought to secrete IL-4

and IFN-y. Their role is yet to be confirmed, however, it is thought they may act as a

precursor cell to T-helper cells yet to have differentiated into either THl or TH2 cells

(Modlin and Nutman, 1993). Several factors can influence the development of the TH

subsets and these include the antigen itself, the antigen concentration, the route of

antigen administration and the antigen presenting cells (Williams et al., 1991;

Gajewski and Fitch, 1991; Bretscher, 1991). THI cells, called inflammatory CD4+ T

cells, are specialised to activate macrophages to have a greatly increased ability to kill

intracellular bacteria. TH2 cells, called helper CD4+ T cells activate antigen-binding

B cells to differentiate into antibody-secreting cells (Janeway and Travers, 1994).

65

Although much of the literature discusses the importance of the humoral aspect of the

immune response in periodontal disease, models used in earlier years suggested that

the initial periodontal lesion is composed mainly of T lymphocytes, with B cells and

plasma cells predominating at a later stage (Mackler et al., 1977; Seymour et al.,

1982; Seymour et al., 1983). Studies have also reported differences in CD4: CD8

ratios in lesions and peripheral blood of periodontitis patients (Okada et al., 1982;

Taubman et al., 1984), compared to healthy controls. This indicates that the cell

mediated immune system is potentially as important in a discussion of this disease as

that of the humoral immune response.

It appears that the immune response in periodontitis takes the following pattern, the

antigen is picked up by an APC, such as a Langerhans cell at the site of infection. It is

then carried to a primary lymphoid tissue where presentation to a circulating naive T

cell takes place. This leads to antigen-specific clonal activation, where some activated

cells become effector cells and others remain in the circulation as memory cells.

Naive and memory T cells may be recognised by the markers they express. Naive T

cells express CD45RA and memory cells CD45RO.

CD45 is a transmembrane tyrosine phosphatase with 3 variable exons that encode part

of its external domain. In naive T cells, high molecular weight isoforms CD45RA are

found. In memory T cells, the variable exons are removed by alternate splicing of

CD45 RNA and this isoform is known as CD45RO. General dogma indicates that

memory T cells (CD45RO+) greatly out number (CD45RA+) T cells in periodontitis

lesions (Gemmel et al., 1992; Yamazaki et al., 1993; Kinane et al., 1999; Lappin et

al., 1999). However, it has been shown that a proportion of these cells are CD45RA+,

which suggests reactivation of the memory cell population by specific but as yet

unidentified antigens in the tissues. It has been reported that memory T cells adhere to

vascular endothelial cells and augment their permeability to macromolecules (Damle

and Doyle, 1990). This functional ability of memory T cells may be a dominant factor

that contributes to the preferential migration of memory T cells into sites of chronic

inflammation (Pitzalis et al., 1988), such as the diseased gingival tissues.

66

The realisation that T lymphocytes are present in large numbers in the diseased tissues

of aggressive periodontitis patients and not in health, has led to investigations of the

numbers, ratios and functionality of T cells in both the peripheral blood and diseased

tissues of patients with periodontal disease. Studies of peripheral blood T lymphocyte

subsets in aggressive periodontal disease have revealed a wide variety of contradictory

results with either depressed CD4+: CD8+ ratios (Kinane et al., 1989b), or a mixture of

increased and depressed proportions of CD4+: CD8+ cells (Katz et al., 1988) or even

no correlation with disease (Engel et al., 1984). Studies have also looked at the

distribution of T cell subsets in tissues. One study reports that the CD4+: CD8+ ratio in

the tissues of aggressive periodontitis patients and even their families tend to be

increased (Syrjanen et al. 1984). A study by Stoufi et al. (1987) and one by Celenligil

et al. (1993), however, clearly states that the ratio is decreased in patients with this

disease. What is obvious, and further indicated by a study investigating juvenile

periodontitis (Meng and Zheng, 1989), is that the ratios are highly variable between

different patients with the same disease, making generalisations and conclusions very

difficult. What the latter study did suggest however, was that high CD4+: CD8+ ratios

were seen in those patients with a greater degree of inflammatory cell infiltration,

while the low CD4+: CD8+ ratio tissues were less inflamed.

When diseased gingival tissue is removed and the cells extracted, both CD4+ and

CD8+ lymphocytes are present in large numbers (Stoufi et al., 1988; Taubman et al.,

1984). However, CD4+ cells although prominent, were found at lower levels than in

the peripheral blood. Overall ratios of CD4+: CD8+ were not found to be altered in

aggressive periodontitis and chronic periodontitis patients (Marthur and Michalowicz,

1997), however, in one study these ratios were found to be depressed in patients with

localised and generalised aggressive periodontitis (Kinane et al., 1989b).

1.4.15 T cell functions

1.4.15.1 The role of CD4+ T cells in periodontal disease

T cell functions in periodontal granulation and gingival tissues can be elucidated by

the cytokine synthesis profile (Yamamura et al., 1991) (see Table 1.2 for

67

characteristics of cytokines that play an important role in the progression of

periodontal disease). Although the cytokine profile is a good tool for T cell subset

investigations, the results reported are often conflicting causing confusion. Often the

results cannot assess the relative importance of the THI and TH2 subsets.

Fujihashi et al. (1996), investigated THl and TH2 cytokine mRNA expression by

CD4+ T cells from diseased gingival tissues. They detected IFN-y, a THI cytokine,

but failed to detect the TH2 cytokine IL-4, thus indicating a more dominant role for

TH I cells. Other groups indicated a more dominant role for TH2 cells after detection

of IL-4, IL-5, IL-6 (Yamazaki et al., 1995; Aoyagi et al., 1995; Manhart et al., 1994)

and IL-10 (Gemmell et al., 1994). One of the current theories of the TH1/TH2

paradigm is that both cells have an important but different role. Gemmell et al.

(1997), have suggested that TH1 cells remain tightly localised at sights undergoing

active destruction. Whereas, TH2 cells are more widely distributed throughout the

tissue and typify a more quiescent stage of the disease. In other words the immune

response in periodontal disease is predominantly TH2 driven with active phases of the

disease leading to foci of TH1 activity.

The role of these cell types is that of help for the immune response. TH1 cells,

characterised often by their production of IFN-y, are important in microbial killing and

are known to augment cytotoxic T cell functions through their production of IL-2 and

IFN-y. TH2 cells, however, inhibit much of the work done by TH1 cells. The

production of IL-4 and IL-10 by these cells inhibits the actions of IFN-y, and hence

has anti-inflammatory characteristics. The production of IL-4, IL-5 and IL-6 elicits

and helps maintain the humoral immune response.

1.4.15.2 Gamma delta T cells

CD4+ and CD8+ T cells usually express the c43 T cell receptor (TCR). In contrast to

this, cells expressing the 78 TCR are typically CD3+and CD4- CD8-. T cells bearing

76 TCRs are a distinct lineage of cells of unknown function. The ligand for this

receptor is also unknown although attempts to identify the ligands have focused on

68

MHC class-I like proteins, mycobacteria, and heat shock proteins (Haas et al., 1990).

There has been growing suspicion over the last decade that 76 T cells do not recognise

antigens in the classic T cell: MHC complex way and that this T cell subset may

recognise antigens presented by non-classical MHC proteins or members of the CDI

family of proteins (Haas et al., 1990).

In peripheral lymphoid tissues, 1-5% of CD3+ cells express the 76 TCR. However, in

epithelial tissues a much higher percentage of the T cells express this receptor

(Janeway and Travers, 1994). yb T cells do not appear to have a great diversity in

their receptors. If 78 T cells found in surface epithelium are examined it can be seen

that the receptors are essentially homogenous in any one epithelium. One explanation

for this is that these T cells are part of the non-adaptive process of the immune

response. Because they have a single receptor and do not re-circulate, could be the

reason why they recognise alterations on the surface of epithelial cells infected with a

pathogen, rather than specific features of the pathogen (Janeway and Travers, 1994).

The role of yS T cells is still in question. They have been reported to lyse various

cellular targets including chicken erythrocytes, cells treated with mycobacteria or

staphylococcal enterotoxin A, and various leukaemic cells (Haas et al., 1990). Data

seems to suggest that these T cells play a protective role in infectious diseases.

Abnormal cell proportions of 76 T cells have been reported in diseases caused by

intracellular pathogens, such as leprosy and leishmania (Marthur and Michalowicz,

1997). High numbers of 76 T cells have been reported in leprosy lesions (Bloom et

al., 1992), rheumatoid arthritis (Reme et al., 1990), Behcet's syndrome (Fortune et al.,

1990), and in the peripheral blood of patients with measles infections (Haas et al.,

1990).

The role of y8 T cells in periodontal disease is still not well understood. It has been

reported that the mean percentage of y6 T cells in the peripheral blood of patients with

periodontal disease, is similar to the numbers found in healthy controls. However,

when the data is observed more closely it can be seen that although the mean values

69

are similar, patients with periodontal disease appear to have abnormally high or low

proportions of 76 cells (Nagai et al., 1993).

y6 T cells have been routinely identified in diseased periodontal tissues (Lundqvist

and Hammarström, 1993; Gemmel and Seymour, 1995; Prabhu et al., 1996). They

have generally been shown to constitute about 1-3% of the CD3+ cells residing in the

diseased tissues (Gemmel and Seymour, 1995; Kawahara et al., 1995). The numbers

of these cells have been shown to increase significantly with disease, however, remain

reasonably low in an environment of moderate to severe inflammation (Gemmel and

Seymour, 1995). y6 T cells from diseased tissues have been shown to express either

CD4 or CD8 (Lundqvist et al., 1994), mRNA for IFN-y, TNF-a, TGF-(3 and IL-6.

However, the cytokine profile is dependent on whether the yb cell expresses CD4 or

CD8.

A number of contradictory studies have been reported on the presence of yE T cells in

healthy tissues. Lundquist and Hammarström (1993), reported that greater than 30%

of all leukocytes separated from healthy tissues were yb+. They described these cells

as being y8+, CD4-, CD8- and CD45RA+, hence naive. Contradictory to this,

Kawahara et al. (1995) reported that when using immunohistochemistry, yb+ cells

were rarely seen in healthy tissues.

The role of yb T cells in periodontal disease remains elusive. One theory suggested

for function of these cells in this disease, is that they may ligate with heat shock

proteins (HSP) produced by stressed or damaged cells. This could be by binding to

and destroying damaged or infected epithelial cells (Lundqvist et al., 1994) or even

stressed bacterial cells. However, whether these cells have a more important role in

the acute rather than the chronic lesion or vice versa is unknown.

1.4.16 Natural killer cells

Natural Killer cells (NK cells) are also said to constitute a large part of the cell

mediated immune system, although they are classed as a component of the innate

70

immune system. These cells are large granular lymphocytes which are often detected

by the marker CDI6. This makes actual numbers detected very difficult to report

since this marker is also found on other peripheral blood lymphocytes and monocytes.

In contrast to CD8+ cytotoxic T cells, which kill infected targets cells in an antigen-

specific manner, NK cells kill infected target cells or tumour cells in an antigen-non-

specific manner (Marthur and Michalowicz, 1997).

The role of NK cells in periodontal disease at present is unknown. However, there are

reports of an increase in the numbers of NK cells in the peripheral blood of patients

with aggressive periodontitis (Celenligil et al., 1990; Watanabe et al., 1993; Kopp,

1988). Conflicting evidence comes from other studies that have reported no increase

in the numbers of NK cells found in the peripheral blood of diseased patients when

compared with health (Zafiropoulos et al., 1990). In the tissues of diseased patients

an increase of NK cells is seen. This is not surprising since in health very few NK

cells are present in the gingival tissues. The numbers of NK cells in the tissues are

seen to increase from health to gingivitis to periodontitis (Wynne et al., 1986; Cobb et

al., 1989; Fujita et al., 1992). However, the proportion of these cells relative to total

lymphocyte numbers actually decreases (Cobb et al., 1989).

The activity of NK cells is increased by soluble mediators such as IL-2, IFN-a, IFN-ß

and IFN-y. The interferons a and ß are antiviral proteins synthesised and released by

leukocytes, fibroblasts and virally infected cells and IL-2 and IFN-y are released by

activated T cells (Benjamini et al., 1996). Surface LPS from Gram negative bacteria

appears to provide the major activation signal for NK-cell-mediated cytotoxicity in

periodontal disease (Lindemann, 1988). This leads to NK cell cytotoxic activity

against host cells.

1.5 Cytokines in Periodontal Disease

1.5.1 Interleukin-1

In the periodontal tissues IL-lß is predominant over IL-la, and compared with

healthy and stable sites is found at higher levels in active disease (Jandinski et al.,

71

1991; Stashenko et al., 1991a). Variations have been noted in the concentrations of

IL-I between sites in the same individual, this coupled with variations in serum IL-lß

levels between chronic periodontitis patients and healthy controls, suggests the

production of this cytokine is at the local level rather than the systemic (Tatakis, 1993;

Chen et al., 1997). Macrophages, PMN, B cells and fibroblasts secrete IL-1(3, and

activated keratinocytes and Langerhans cells produce both IL-Ia and IL-I P (Hillmann

et al., 1995; Jandinski et al., 1991; Gemmel and Seymour, 1998; Takahashi et al.,

1995).

The production of IL-1 stimulates certain cells to express adhesion molecules. These

include; fibroblasts, endothelial cells, monocytes and granulocytes (Takahashi et al.,

1994). The expression of adhesion molecules aids the migration of immune cells

from the capillaries into the inflamed tissues. It has been shown that numerous

periodontal pathogens can stimulate host cells to produce IL-1 (Gemmel et al., 1992;

Gemmel and Seymour, 1998).

IL-I can induce the production of plasminogen activator, IL-6, matrix

metalloproteinases and PGE2 by gingival fibroblasts. These cells are involved in

tissue destruction, turnover and remodelling (Richards and Rutherford, 1988). IL-1

stimulates osteoclasts to resorb bone (Horton et al., 1972), whilst inhibiting bone

formation (Stashenko et al., 1987). IL-1-mediated bone resorption can be induced by

periodontopathogens (Ishihara et al., 1991).

1.5.2 Tumour necrosis factor

TNF-a, IL-la and IL-lß are all known as pro-inflammatory cytokines and have all

been identified at higher levels in periodontitis lesions from patients with chronic

periodontitis compared with healthy sites (Matsuki et al., 1992; Stashenko et al.,

1991b). However, cells containing ILA P were found in much greater numbers than

those producing TNFa and IL-la (Stashenko et al., 1991b). In contrast to IL-1, only

very low concentrations of TNFa have been demonstrated in GCF from both chronic

72

and aggressive periodontitis patients (Rossomando et al., 1990; Yavuzyilmaz et al.,

1995).

High levels of TNFa in periodontal lesions appear to represent an established

inflammatory and immune response. Salvi and co-workers (1998) examined GCF IL-

1(3, TNFa and PGE2 levels in type I diabetes mellitus patients with periodontal

disease. They found that diabetics had significantly increased levels of PGE2 and IL-

113 but not TNFc , as compared with non-diabetic controls with similar periodontal

status. In addition, diabetics with moderate to severe disease had almost two-fold

higher levels of PGE2 and IL- lß compared with diabetics suffering from gingivitis or

mild periodontitis.

PGE2 causes increased vascular dilation and permeability. It also stimulates

macrophages to secrete matrix metalloproteinases and has been shown to trigger bone

resorption in vitro (Birkedal-Hansen, 1993), thus acting synergistically with IL-1 and

TNFa. Furthermore it upregulates complement and Fe receptors on the surface of

monocytes and PMN (Offenbacher, 1996). In inflamed periodontal tissue,

prostaglandins and leukotrienes are mainly produced by activated macrophages,

although they can also be produced by fibroblasts. Prostaglandins, especially PGE2,

comprise the primary pathway of alveolar bone destruction in periodontitis.

Leukotrienes, especially leukotriene B4, are potent chemoattractants for PMN.

1.5.3 Interleukin-10

A role for IL-10 in the progression of chronic periodontitis was suggested by Gemmell

and Seymour (1998). This study indicated that T cell lines from both the peripheral

blood of periodontal patients and their gingival tissue, secreted IL-10, but clones

derived from the peripheral blood of an individual with gingivitis did not. It has also

been suggested by Stein and co-workers that IL-10 may contribute to auto-immune

reactions against gingival tissues (Stein and Hendrix, 1996; Stein et al., 1997).

Gemmell and Seymour (1998) have demonstrated a significantly reduced number of

IL-10+ CD8+ cells from chronic periodontitis lesions, than from healthy/gingivitis

sections. They suggested that in gingivitis, IL-10 might suppress inflammation by

73

decreasing macrophage activity, thereby preventing progression to periodontitis. No

differences were found in the percentage of 1L-10+ CD4+ cells between the two

disease categories. However, some individuals were found to have higher numbers of

IL10+ T cells, regardless of disease category. It appears that the overall variation

between the two diseases entities in IL-10+ CD8+ cells may have been due to

individual variation in IL-10 secretion patterns, rather than differences in pathogenesis

between the two disease categories. Further studies are needed to clarify the role of

IL-10 in periodontal disease.

In summary, IL-1, IL-6 and TNF are pro-inflammatory cytokines which are detected

and may be modulated within the periodontium. They are important aspects of the

host defence system, operating in the periodontal tissues during health, destruction

and healing phases. Further research may lead to a greater understanding of the roles

of other members of the cytokine network, in particular IL-10 and various soluble and

cell-bound receptors and inhibitors of cytokine function, in periodontal disease. These

molecules are inter-related and are part of the immune, inflammatory, breakdown and

repair homeostasis ongoing within the periodontium.

74

Table 1.2 The Role of Cytokines in Periodontal Disease

Cytokine Produced by Role Interleukin-1 (IL-1) large quantities " proinflammatory properties

produced by " mediator of tissue destruction in macrophages periodontal disease (Alexander &

Damoulis; 1994) Interleukin-4 (IL-4) activated T cells " inhibits production of IL-I, TNF

and IL-6 " an absence of IL-4 in the

periodontal tissues has been suggested to trigger disease progression (Fujihashi et al., 1993, Hirano et al., 1991)

Interleukin-6 (IL-6) lymphoid and " mediates inflammatory tissue non-lymphoid cell destruction types. " thought to be an important Production is cytokine in B cell differentiation, triggered by IL-1, and hence play a role in the TNF and IFN-y induction of the elevated B cell

response in patients with periodontal disease (Fujihashi et al., 1993)

Interleukin-8 (IL-8) mononuclear " attracts and activates neutrophils monocytes and into the tissues of the many of the tissue periodontium, particularly as it is cells produced by gingival fibroblasts

(Takashiba et al., 1992) Interleukin-12 (IL-12) monocytes, " Pleiotrophic effects on NK cells

macrophages, B and T cells in the tissues cells and " induces IFN-y production and is accessory cells necessary for THI induction

Tumor Necrosis Factor a macrophages and " similar effects to IL-1 and IL-6

and P (TNF-a, TNF-ß) lymphocytes respectively

Interferon y (IFN-y) activated T cells " potent inhibitor of IL-1, TNF-a I and TNF-

75

1.6 Important Antigens in Periodontal Disease

1.6.1 Introduction

The following section is a discussion of immunodominance and the existence of

immunodominant antigens. The aim of the studies presented in this thesis is to focus

on the target of the humoral immune response in periodontal disease, and in particular

the importance of immune response inducing antigens. The subject of

immunodominance is a controversial one, and the following section will be devoted to

discussing evidence in the literature for immunodominant antigens in a number of

diseases.

There is substantial literature on the humoral immune response detected against

micro-organisms present within periodontal pockets. However, the significance and

specificity of the immune response in periodontal disease has proved difficult to

elucidate. This is due to the large number of putative pathogens in the plaque biofilm,

and the apparent commensal nature of many of these opportunistic organisms.

Identification of the immunodominant antigens involved in the disease process would

be informative especially with respect to: (i) the relative importance of the implicated

pathogens; (ii) new approaches to immunological diagnosis; (iii) specific bacterial

virulence determinants; (iv) natural protective responses; and (v) the selection of

potential candidate antigens for vaccine development.

The concept of immunodominance is often misused by investigators. Their

interpretation of data is that specific antigens or epitopes of a micro-organism or

macromolecule are ìmmunodominant'. However, little explanation as to the basis of

the finding and importantly, its significance is provided.

The phenomenon of ìmmunodominance' was initially described by Sercarz and

colleagues (Sercarz, 1989) to provide a framework for explaining the genetic control

of antibody responses to hen egg lysozyme (HEL). HEL is a globular protein with a

known 3-dimensional structure. These investigators were able to define antigenic

epitopes using peptides derived from cyanogen bromide cleavage of this protein.

76

Specifically, immune responses of H-2e mice to HEL were restricted to 3 amino

terminal residues of the molecule. Wicker et al. (1984), reported that if these 3

residues were cleaved, 50% of the primary response antibody molecules bound poorly

if at all to the HEL, and the immunogenicity of the protein was drastically effected.

The term ìmmunodominance' was used to describe the predominant immune

recognition of these 3 amino terminal residues. Sercarz (1989) suggested that for

efficient and regulated activity of the immune response to a particular antigen, there

should be a small number of immunodominant epitopes on an antigen.

The primary, secondary, and tertiary structures of a protein antigen offer a vast array

of potential epitopes. However, the number recognised by the immune system

appears to be considerably smaller than the full antigenic repertoire of the immunogen

(Laver et al., 1990). Some of these antigenic epitopes will dominate the resulting

specificity of the immune response. This can vary from species to species, and even

among individuals within a species. They will not only induce a more pronounced

immune response, but may also suppress the primary immune response to other

antigenic determinants. Recent studies using a similar experimental system have

demonstrated the broad heterogeneity of proliferative responses to the

immunodominant determinants within HEL. The heterogeneity of response was

suggested to relate to the competitive and regulatory nature of the interaction between

various factors. These include different MHC molecules, determinant capture by

antigen presenting mechanisms, and the available T cell repertoire. The authors

suggested that these results may have important implications for studies of

autoimmunity, infection and vaccine design in human populations, where

heterozygosity is the characteristic of the population (Moudgil et al., 1998).

The basis for the existence of immunodominant epitopes is not well understood.

Recent findings have shown that peptide competition for MHC binding may not

represent the most important event in processes leading to immunodominance (Lo-

Man et al., 1998). T lymphocyte responses to a protein antigen are restricted to a

limited number of determinants and not to all peptides capable of binding to MHC

class 11 molecules.

77

Focusing of the T cell immune response is also defined as immunodominance, and has

been observed with numerous antigens. Recent investigations by Gapin et al. (1998)

have suggested that dendritic cells play a major role in the immune response against a

few antigenic determinants. Conversely, B lymphocytes may diversify the T cell

response by presenting a more heterogeneous set of peptide-MHC complexes.

Moreover, this T cell focusing has been extended to cellular immune responses, which

are directed against a narrow set of immunodominant peptides derived from complex

antigens. The existence of epitopes that are hidden or infrequently targeted by

immune responses (cryptic epitopes) has also been identified.

Although the identification of immunodominant epitopes is important for vaccine

development, understanding immunological reactivity to these cryptic epitopes may

be important in the development of autoimmunity. Blum et al. (1997) have suggested

that targeting exogenous antigens into specialised processing compartments within

antigen presenting cells appears to be important in epitope selection and

immunodominance.

The final outcome of the immune response may be an immunochemical phenomenon

which is related to the protein structure or the host's immunoregulatory mechanisms.

However, it also appears to be effectively used by various pathogens, as a host evasion

tactic. This occurs when the microbial immunodominant epitopes are sterically close

to antigens which have the capacity to demonstrate antigenic diversity and are capable

of antigenic variation. Thus, they help to conceal the micro-organism from the host

protective responses (Nara et al., 1998).

In the field of periodontal research the term ìmmunodominant antigen' and the

concept of ìmmunodominance' have been used imprecisely, are poorly defined and

often misrepresented. There are multiple reports in the literature, purportedly

characterising immunodominant antigens of oral micro-organisms, yet no clear

definition of immunodominance is included. It is not known whether these antigens

express an epitope that is the focus of the primary immune response, and which is

essential for the initiation of the response to the antigen and micro-organism (Wicker

78

et al., 1984). Due to the chronic nature of periodontal infections, these subjective

measures may only be defining the gravimetric quantity of a particular antigen, rather

than describing a function of its unique antigenicity.

Thus, the original definition of immunodominance has been distorted to describe a

microbial antigen to which a predominance of antibody is detected when compared to

the myriad of other antigens presented by the micro-organism and recognised by the

host. Investigators should be clearer when reporting results. They should indicate

whether they have described an antigen to which there is a `detectable antibody

response' rather than an ìmmunodominant antigen'.

Immunodominance and immunodominant antigens have been difficult to study in

periodontal disease due to the large number of species comprising the microbial

biofilm at sites of health and disease. Other factors including proposed genetic

predisposition, behavioural variables such as oral hygiene levels, and environmental

factors (e. g. smoking and socio-economic factors) have not made the problem any

simpler. In addition, there is controversy about the nature of the humoral immune

response itself. Whether the antibody response in this disease is protective or

contributes to tissue destruction indirectly is unknown. It is reasonable to assume that

disease activity would be abrogated if the antibody response was effective at

eliminating or controlling the growth of the micro-organisms present. In contrast the

antibody response may be harmful to the host if it protects certain bacteria or actually

causes pathology itself (e. g. hypersensitivity). A study by Mooney et al. (1995),

reported that patients with periodontal disease who possessed high avidity antibodies

to oral pathogens, have a better prognosis than patients with low avidity antibodies,

regardless of whether the latter have a high titre of antibodies or not. This study

appears to indicate that an effective antibody response is protective. However,

numerous reports speculate on a possible autoimmune aspect to periodontal disease

(Brantzaeg and Tolo, 1977; Kristofferson and Tonder, 1973; Hirsch et al., 1988;

Anusakasathien and Dolby, 1991). There is no conclusive evidence that high levels of

antibody are protective or indicative of disease. An important factor which must be

considered is that, patients with periodontal disease are not found to be more

susceptible to other bacterial infections. They also do not show severe pathology

79

linked to any other diseases. In addition, autoantibodies have not been found to

predominate in the periodontal lesion, although antibodies to host components, in

particular collagen (Ftis et at., 1986; Peng, 1988; Hirsch et at., 1988) have been

reported. Until evidence is found to support the conclusion that the immune response

is destructive it should be assumed that it is protective. However, the pathology seen

in this disease is not only due to the direct harmful effects of micro-organisms.

Production of high levels of pro-inflammatory cytokines (IL-6, IL-8, TNF) has been

reported, (Tsai et at., 1995; Yavuzyilmaz et al., 1995) which could result in

considerable tissue destruction.

A further consideration in periodontal disease, is that critical clinical features of the

disease are often identified long after the biologic mechanisms have initiated the

disease process.

When addressing our current knowledge of the antigens to which a detectable immune

response in periodontal disease may be directed, one must consider that within a

complex microbial biofilm, such as the subgingival plaque, it appears that certain

micro-organisms elicit greater antibody levels (i. e. dominate the immune response)

than others (Zambon 1996). This characteristic of the host response does not appear

to depend entirely on the absolute numbers of an individual micro-organism. It may

be related more to their structural and antigenic components or their pathogenic

significance (virulence), as well as their distribution within the biofilm and/or host

tissues. An understanding of the antigens, which elicit a detectable antibody response,

should contribute to a better understanding of specific bacterial virulence

determinants, natural protective responses, and even possible strategies for vaccine

development, or new approaches to immunological screening and diagnosis of the

disease.

A survey of the scientific literature on immunodominant antigens in medical

infectious diseases provides an indication of the significance of this concept for

identifying potential virulence, diagnostic, or vaccine candidates. In the following

discussion, selected studies are reviewed to provide examples of the variation in the

use of the terminology. In addition, the potential limitations of equating detectable

80

antibody response with immunodominant antigen and protective immune responses

are highlighted.

Examples of reports identifying immunodominant and therefore, candidate diagnostic

antigens in parasitic infections include studies on Neospora canium (Howe et al.,

1998), Theilera pat-va (Katende et al., 1998), Trypanasonia cruzi (Pereira et al., 1998)

and Schistosomna inansoni (Chen and Boros 1998). It is clear from the literature that

an antigenic relationship to similar antigens from other parasites and prevalence of the

response during infection, is often used as evidence to speculate on the importance

and function of potentially "important" antigens during infection. Identification of

important antigens from these types of studies can also lead to a higher degree of

sensitivity and specificity in diagnosis. Some studies have even uncovered

information on the disease pathology and progression. For example, Chen and Boros

(1998) identified that the immunodominant T cell epitope, P38, of Schistosoina

mansoni, elicited the pulmonary granuloma formation, the disease associated

hypersensitivity lesion of this infection.

There have also been a variety of similar studies which focused on dominance in

bacterial or fungal infections. While the mortality rate for systemic candidosis

remains high, the diagnosis is somewhat difficult and there has been considerable

interest in developing a reliable serodiagnostic test. Mathews et al. (1988) identified

the 47 kDa component of Candida albicans as an ìmmunodominant antigen' in the

serology of systemic candidosis. The 47 kDa antigen appears to be immunodominant

and 92% of patients with candidal antibodies produced a response to an antigen of 47

kDa. It seems that this antigen is conserved in structure between strains, and an assay

detecting this antigen would be the basis for a sensitive and specific diagnostic test

(Mathews et al., 1984,1987).

Further interesting and important knowledge has been gained from studies involving

immunodominant antigens and the quest to identify them. The Mycobacterium avium

complex (MAC) consists predominantly of Mycobacterium avium (M. avium) and

Mycobacterium intracellulare (Inderlied et al., 1993). MAC members are ubiquitous

environmental micro-organisms and infection in healthy individuals is rare. However,

81

MAC infection is a major cause of bacterial morbidity and mortality in AIDS patients.

A recent report by Triccas et al. (1998) suggested, following molecular and

immunological analyses, that a 35 kDa protein of M. aviuni is a homologue of the 35

kDa immunodominant antigen of Mycobacterium leprae (Triccas et al., 1996). The

35 kDa protein of MAC, was confirmed to contain T-cell and B-cell epitopes

recognised by human immune responses following mycobacterial infection. However,

it was discovered that immunologically there is an inability to distinguish between

MAC and Mycobacterium tuberculosis, even though the 35 kDa protein is not

recognised by tuberculosis patients. Although this is an example of a study where an

antigen eliciting detectable antibody response is identified, it does not identify a

potential vaccine candidate or any other markedly positive conclusion. For both

tuberculosis and mycobacterial infections, early diagnosis is crucial for the prognosis

of the patients. Therefore, this emphasises a situation where using a detectable

antibody response for differential diagnosis is of critical importance.

Studies investigating dominant antigens, even if they are not primarily designed to do

so, can answer crucial questions. Otitis media is a common problem associated with

high morbidity, in which a subset of children experience repeated episodes caused by

nontypeable Haentophilus influenzae (H. influenzae) (NTHI). Murphy and

Kyungcheol (1997) showed that the principal antibody response in animals to an

NTHI strain was directed at an immunodominant epitope of the P2 molecule, which is

the major outer-membrane protein of NTHI. However, this group hypothesised that

the otitis prone children develop an antibody response to an immunodominant region

of the P2 molecule, which is strain specific. Therefore, the child's protective response

is directed exclusively against the infecting strain. These children will experience

recurrent episodes of otitis media since they have no protection against a new strain.

The discovery of the dominant antigen associated with this infection has further

important implications. It contributes to the understanding of the recurrent nature of

otitis media and provides useful data on a potential vaccine design for this disease.

As mentioned in the preceding section, the importance of T cell immunodominant

epitopes is clear and must be distinguished from those eliciting a dominant B cell

response. Listeria monocytogenes secretes proteins associated with its virulence

82

inside infected cells. These are degraded by host cell proteasomes, processed into

peptides and bound to MHC class I molecules. These antigen epitopes are effectively

presented to cytolytic T lymphocytes (CTL). Results of Pamer et al. (1997) indicated

that immunodominant T cell responses cannot be predicted by the prevalence of

antigens or epitopes. In fact, one of the least prevalent epitopes primes for the

immunodominant CTL response in mice.

Above are all examples of the purported significance of identification of

immunodominant antigens and the potential benefits in medically important

infections. Utilisation of the antigen(s) to create diagnostic tools, and develop

vaccines attaches greater value and relevance to the search for these types of antigen.

The need to identify periodontal pathogens as true aetiological agents in the disease

process still exist. Many of these candidate pathogens may be merely opportunistic

and depend on the biofilm "food web" to create the conditions or niche within which

they can colonise, obtain nutrients and replicate. Thus, these micro-organisms would

not classically be considered as the true pathogens against which host defence

mechanisms should be directed. In spite of the potential microbial complexity of this

disease, a number of candidate species have emerged which are thought to play a role

in the pathogenesis of periodontal disease. These micro-organisms generally fulfil the

modified Koch's postulates as described by Socransky and Haffajee (1992).

Examination of numerous periodontal pathogens has revealed these micro-organisms

to possess a variety of potential virulence factors and structures that may be crucial to

their pathogenicity. In the quest for the identification of the important bacterial

antigens in periodontal disease, a number of different candidate antigens have been

suggested.

1.6.2 Outer-membrane proteins

Much of the periodontal literature applicable to this topic covers studies on the outer-

membrane proteins (OMP) of periodontal pathogens. OMP of Gram negative bacteria

may reside on the inner or outer layer, or can span the entire outer-membrane. Gram

83

negative bacteria possess multiple OMP which provide important biological functions,

such as phage receptors or receptors for uptake of nutritional substrates. Various

specific microbial antigens have been identified from the outer-membranes of oral

pathogens to play a role in periodontal disease. However, many studies have

identified outer-membrane antigens of the same micro-organisms, which remain

uncharacterised and may be of importance in protection afforded by the adaptive host

immune response (Watanabe et al., 1989; Califano et al., 1989; Wilson et al., 1991;

Califano et al., 1992; Page et al., 1992b). These studies have tended to focus on

either sonicate antigens or OMP which can be sheared from the surface of the bacteria

(Nakagawa et al., 1995). Since these two preparation methods differ, one could

identify both antibodies to intracellular antigens, as well as to the surface antigens,

however, similar results might be expected. It seems unlikely that patients would

produce an array of antibodies to "inaccessible" intracellular proteins, and more likely

that antigens eliciting a detectable antibody response would be expressed on the

microbial surface or as secreted proteins. The majority of the outer-membrane

components of putative periodontopathogens remain uncharacterised with respect to

their metabolic or structural functions or their activity as virulence determinants. In

general, patients with strong responses to these individual antigens often tend to be

those who have a strong response to the whole bacterial cells.

Nakagawa et al. (1995) reported detection of antibodies to apparent

ìmmunodominant antigens' of P. gingivalis in the sera of chronic periodontitis

patients. These were specific for antigens of molecular weight 75 and 31 kDa from

outer-membrane preparations, and a 46 kDa antigen from sonicated extracts of the

micro-organism. Further studies identified 75,55, and 43 kDa bands in whole cell

sonicates of P. gingivalis to be ìmmunodominant antigens' in rapidly progressive

periodontitis patients (Chen et al., 1995). A similar approach was utilised by Ebersole

and Steffen (1994). They documented ìmmunodominant antigens' in outer-

membrane preparations, from various strains of P. gingivalis. The results identified

some similarities in antibody recognition of antigen bands across strains, however, a

striking variability was noticed between the strains. Additionally, variations in

response patterns were noted to be associated with high and low response patients,

irrespective of the clinical disease classification.

84

Ebersole et al. (1995) and others (Bolstad et al., 1990; Watanabe et al., 1989; Sims et

al., 1991) have used Western immunoblot approaches to detect antibody responses

specific for antigens in the outer-membranes of A. actinomycete? ncomitans. These

studies revealed a wide variation in responses among individual patients, thus adding

to the difficulty regarding the identification and/or importance of an antigen(s) that

dominates throughout a patient population. Alternatively, Flemmig et al. (1996)

demonstrated IgG/IgA antibody activity in aggressive periodontitis (65%/70%) and

chronic periodontitis (45%/55%) patients reactive with a 110 kDa OMP of A.

actinomycetemcomitans. In contrast, control subjects had no IgG and only 5% had

any IgA antibody activity. This approach exemplifies a strategy whereby a high

frequency and or level of antibody is detected to a particular antigen in patients versus

controls, which is then suggested to be an immunodominant antigen. Each of these

studies aimed to identify immunodominant antigens as macromolecules which may

play an important role in pathogenesis, and the response to which correlated with

disease resistance. In fairness to the authors, many of these studies probably identified

the molecular weight of antigens which elicited a detectable antibody response in this

disease. However, none of the mentioned studies obtained sufficient evidence to

describe an immunodominant antigen.

Importantly, the OMP are uniformly immunogenic and many demonstrate antigenic

diversity (heterogeneity) (Holt and Bramanti, 1991; Holt and Ebersole, 1991; Mitchel,

1991). Moreover, environmental changes altering the growth of numerous micro-

organisms, such as Bordetella pertussis, have been shown to result in antigenic

modulation of OMP (Robinson et al., 1986). Genco and colleagues (1991) presented

the concept of the potential importance of antigenic heterogeneity, as a virulence

strategy for oral micro-organisms. Recent studies by Ebersole and colleagues (1995)

with A. actinomycetemcomitans, Chen et al. (1995) with P. gingivalis, and Sims et al.

(1998) with B. forsythus have supported antigenic diversity of these pathogens within

infected patients. However, there remains minimal information on its potential role in

virulence within the periodontal microbiota.

85

The previously mentioned studies give examples of strategies which have been used

and provide examples of some potentially interesting targets. However, they only

"scratch the surface" in providing an understanding of the antigenic epitopes which

activate the regulatory T cells. These T cells effectively stimulate regional B

lymphocytes which could provide a functional host immunity in periodontal disease.

Current genome sequencing of these pathogens will allow us to visualise the

organisation of these protein genes. In addition it will enhance studies of genetic and

antigenic heterogeneity within the population. Sequencing, purification and

functional studies of these antigens will encourage evaluation of their significance in

well characterised patient and control populations, as well as in in vitro and in vivo

models of disease. More importantly, from the perspective of this review it is not

clear that the concept of a "dominant antigen" is critical to the ultimate goals of this

research. As mentioned previously, high antibody levels or raised titres to an antigen

do not necessarily define immunodominance of the antigen. They may instead

delineate a dominant antibody response. If this response is directed against a `flak

antigen', (an antigen capable of significant diversity or variation, or having little

contribution to the pathogenicity of the micro-organism), the detection of a dominant

antibody response does not inherently describe an immunodominant antigen nor

provide insight into immune protective responses.

1.6.3 Fimbriae

Fimbriae or pili as they are often termed because of their functional and molecular

homology to pap pili of E. coli (Krogfelt, 1991), are curled single stranded filaments

with a diameter of approximately 5nm (Ishikawa et al., 1997). They are cell surface

structures important in adherence and are possibly virulence factors in medically

significant pathogens.

Gilsdorf et al. (1989) reported the fimbriae of H. influenzae as being immunogenic.

Pichichero et al. (1982) detected anti-pilin antibodies in human serum following H.

influenzae infection. However, many investigators have suggested that fimbriae of H.

influenzae are not pathogenic when tested in animal models (Kaplan et al., 1983; Stull

et al., 1984; Mylotte et al., 1985).

86

Early ultrastructural studies clearly identified fimbriae-like structures on the surface of

numerous oral micro-organisms, including A. actinonrycetenicomrtans, Actinomyces

spp., P. gingivalis and Prevotella spp. (Cisar et al., 1979; Woo et al., 1979;

Scannipieco et al., 1983; Rosan et al., 1988; Leung et al., 1989). Therefore, as

identified for other pathogens, these structures may be of importance in the

pathogenesis of periodontal disease. The principle work in this area has involved

detailed studies of these structures in P. gingivalis.

P. gingivalis has been identified in gingival tissues, which suggests that this micro-

organism is able to successfully colonise the microbial biofilm in the gingival sulcus

and cross the epithelial barrier into deeper tissues (Gibson et al., 1964; Saglie et al.,

1988; Sandros et al., 1996). Fimbriae appear crucial for the adherence of micro-

organisms, such as P. gingivalis, to other members of the microbial plaque and to host

gingival tissue surfaces. Supportive evidence arises from studies examining

afimbriate variants of P. gingivalis with diminished capacity to adhere to epithelial

cells (Genco et al., 1994). Hamada et al. (1994) reported that a fimA mutant of P.

gingivalis, which lacked surface fimbriae, showed deficient adherence to both

cultured human gingival fibroblasts and epithelial cells. They concluded that this

protein structure is essential for the micro-organism to interact with gingival tissues

(Hamada et al., 1994).

The functional importance of fimbriae has further been supported by Madianos et al.

(1997), who described the adhesive and invasive properties of P. gingivalis strains on

the KB oral epithelial cell line. In this study, a population of a fimbriated strain of P.

gingivalis demonstrated 50% adherence to the epithelial cells. In addition 10% of the

bacteria were internalised within 90 minutes of infection of the monolayer. In

contrast, a mutant strain lacking fimbriae displayed a significantly decreased adherent

capacity (10%), only 1.5% of the bacteria were internalised. The available literature

supports the theory that fimbriae of this pathogen are a target of the humoral immune

response in periodontal patients. However, a lack of antibodies reactive with fimbriae

has not been identified as increasing disease risk. Condorelli et al. (1998) recently

87

reported detection of IgA antibodies in the GCF of patients with acute recurrent

periodontitis. The antibodies were specific for the purified 43 kDa fimbrial antigen.

Malek et al. (1994) used P. gingivalis with an insertional mutation in the gene

encoding the fimbrillin subunit. They showed that the mutant was unable to adhere to

salivary pellicle coated tooth surfaces and could not induce bone loss in a gnotobiotic

rat model. Additionally, an anti-fimbrial monoclonal antibody has been shown to

block the adherence of P. gingivalis to salivary pellicle-coated tooth surfaces (Genco

et al., 1994). These two experiments suggest that an absence of antibodies to fimbriae

may encourage the adherence/colonisation of P. gingivalis and enhance the risk of

disease onset.

There is generally no evidence supporting the role of fimbriae as an immunodominant

antigen in periodontal disease, or other diseases. These studies have described

dominant humoral immune responses. They do not establish that fimbriae are

immunodominant antigens. Fimbriae seem to be extremely important in mediating

adherence to human mucosal surfaces (Beachey, 1981). Therefore, they must be

considered an attractive candidate for a future vaccine. A study by Brant et al. (1995)

reported a novel approach to induce protection against different oral micro-organisms,

that was to define the immunodominant region of fimbrillin and then use synthetic

peptides to mimic this epitope in a vaccine. This was a novel and theoretically

excellent approach. However, with hindsight the lack of reports in dental literature

about human antibody titres directed against fimbriae, coupled with the difficulty of

extrapolating results from animal models, indicate that there are problems with this

concept. The utilisation of adhesins as vaccine candidates is not a new idea. Clark et

al., (1985) developed a model to evaluate the role of fimbriae in the prevention of

colonisation of teeth by Actinomyces viscosus (A. viscosus). After mice had been

immunised with both type 1 and type 2 fimbriae, antibodies of isotype IgG and IgA

were detected. Further challenge with A. viscosus resulted in a significant reduction in

the number of mice colonised. The results of this early study, indicated that fimbrial

vaccination may modulate colonisation of oral bacteria.

88

In conclusion fimbriae do not appear to be immunodominant although they do appear

to be important. The available literature supports the theory that fimbriae of P.

gingivalis are a target of the humoral immune response in periodontal disease patients.

This has been delineated in various disease populations, with a general finding of

increased levels and frequency of serum antibody to the fimbriae or fimbrillin subunits

(Okuda and Takazoe, 1988; Ogawa et al., 1989a; Genco et al., 1991). Recently,

Condorelli et al. (1998) reported the detection of IgA antibodies in the GCF of

patients with acute recurrent periodontitis. The antibodies were specific for the

purified 43 kDa fimbrial antigen. Thus, it is clear that both a local and systemic

humoral immune response is generated to this antigen in many periodontitis patients.

However, evidence that fimbriae are immunodominant antigens in periodontal disease

or in other infectious diseases is generally lacking.

1.6.4 Lipopolysaccharide (LPS)

A major structural feature of all Gram negative bacteria is the LPS macromolecule

comprising the outer-membrane of these prokaryotes. LPS exhibits a broad spectrum

of pathophysiological properties. It is a potent immunostimulator and adjuvant, as

well as being a principal virulence determinant of most Gram negative pathogens. It

elicits these effects via induction of local and systemic inflammatory reactions.

The LPS macromolecule is composed of 3 distinct regions; the 0 polysaccharide-

chain, the core region, and the lipid part, termed lipid-A. The O-Chain anchors LPS

to the bacterial membrane (Luderitz et al., 1982). This portion of the molecule shows

enormous structural variability within prokaryotes. It is generally characteristic for a

Gram negative bacterial species and often represents an antigenically diverse surface

component of these micro-organisms (Zahringer et al., 1994). The core region of LPS

is organised into an outer core and an inner core (Holst et al., 1992). The outer core

region contains neutral sugars and the inner core often contains rare glycose residues.

Moderate inter-bacterial variability has been suggested to occur in this region. The

lipid A portion of LPS is linked to the polysaccharide core of the molecule. It is the

least variable component of LPS (Rietschel et al., 1984) and is generally considered to

be responsible for the endotoxic, inflammatory and tissue destructive effects of LPS.

89

LPS from oral micro-organisms can penetrate calcified tissue (Aleo et al., 1974) and

gingival connective tissue (Aleo et al., 1980). Studies have also suggested that LPS

from oral bacteria may contribute directly to tissue damage and bone resorption

characteristic of periodontal disease (Hausmann et al., 1970). This early work and

subsequent studies are based on in vitro cell cultures and therefore, may not

necessarily apply to the in vivo situation. Nevertheless, proposing a pathogenic role

for LPS is reasonable since it has been detected in periodontal tissues, even in the

absence of identifiable bacteria. Large numbers of blebs (e. g. outer-membrane

vesicles) have been observed associated with several Gram negative oral bacteria.

These membrane bound, diffusable liposomal-like structures could enhance the

transfer of LPS to sites distant from the plaque biofilm, such as the periodontal tissues

(Nisengard and Newman, 1994).

Numerous studies have demonstrated that the serotype determinants of A.

actinomycetemcomitans are associated with a high molecular weight LPS-associated

antigen (Califano et al., 1989) or carbohydrate smear antigen (Califano et al., 1989;

Sims et al., 1991). In aggressive periodontitis patients with the highest serum

antibody levels to A. actinomycetemcomitans the predominant antibody specificity is

to LPS (Okuda et al., 1986; Wilson et al., 1991; Califano et al., 1992; Wilson and

Hamilton, 1995). LPS is generally thought to exhibit immune stimulating

characteristics which are T cell independent, and thus principally induce an IgM

response. However, it is clear that aggressive periodontitis patients produce

substantial levels of IgG antibody to this antigen from A. actinomycetemcomitans (Lu

et al., 1993). The level and characteristics of this dominant antibody response

suggests that an anti-LPS response is a significant marker of A.

actinoniycetemcomitans infection and existing periodontitis. However, whether LPS

represents a virulence determinant, a true "dominant antigen", or an antigen to which

the adaptive immune response can provide protective immunity remains to be proved.

The literature is replete with studies reporting high titres of serum IgG antibodies to P.

gingivalis in periodontitis patients compared with healthy controls. However, the

temporal acquisition of the antibody response relative to colonisation and/or infection

90

has not been clearly described. Thus, animal models have been developed to examine

the acquisition and specificity of potential protective immunity. A recent study by

Vasel et al. (1996) reported that immunisation of monkeys with a vaccine containing a

monkey isolate of P. gingivalis, induced protection against alveolar bone destruction.

This might be considered surprising since a number of pathogens are considered to be

associated with the periodontopathic microbial ecology (Socransky and Haffajee,

1992; Zambon, 1996). In addition, in many periodontitis sites in humans, P.

gingivalis is undetectable (Mombelli et al., 1991). Further investigation of immune

and pre-immune sera obtained from monkeys, used immunological analyses including

ELISA and Western immunoblotting (Vasel et al., 1996). After immunisation the sera

of the monkeys contained increased antibody levels reactive not only with antigens of

P. gingivalis, but also with antigens of B. forsythus. Detailed analyses suggested that

the binding of serum antibodies was a specific, adaptive response to the LPS in the

immunogen, since little reactivity was detected in the pre-immune sera. Thus, the

antibodies induced by LPS from P. gingivalis appeared to be cross-reactive with the

LPS of B. forsythus, and could contribute to a more general modification of plaque

ecology.

These observations are similar to those obtained by other investigators examining LPS

in non-oral Gram negative bacteria. For example, Seifert et al. (1996) reported the

generation of a human monoclonal antibody that recognised a conserved epitope

shared by LPS molecules of different Gram negative bacteria. This indicated a high

degree of cross-reactivity of antibody across different bacterial species. The epitope

recognised by the human monoclonal IgM antibody (i. e. LPD5H4) was located in the

lipid A portion of the LPS molecule. Importantly, the monoclonal IgM antibody was

shown to partially neutralise LPS activities of Salmonella minnesota, Salmonella

typhimurium, E. coli, Klebsiella pneumoniae, Pseudomonas aeruginosa and Neisseria

meningitidis. The binding specificity of LPD5H4 is only a small step in the search for

the existence of immunodominant epitopes in the conserved regions of LPS.

However, it does provide an indication of the potential for broad cross-reactivity of

antibodies against LPS molecules of different bacterial genera. This phenomenon

should be considered when documenting the role of LPS as an antigen in periodontal

disease.

91

Preston et al. (1996) have reported that although many would assume LPS to be the

major glycolipid in the cell wall of Gram negative bacteria, in fact lipooligosaccharide

(LOS) is. LOS and LPS both share lipid A in their structure (Griffiss et al., 1995).

Therefore, if lipid A is the immunodominant portion of these glycolipids it could

result in even broader cross-reactivity. Nevertheless, polyclonal anti-E. coli lipid A

antibodies have been shown to cross-react with a large variety of free lipid A

molecules. However, antibodies to lipid A do not cross-react with intact LPS in vitro,

indicating that either the cross-reacting lipid A epitopes are not expressed or that the

epitopes are cryptic (Zahringer et al., 1994). These reports often evaluated

monoclonal antibodies of murine or human lineage. A cautionary note was proved by

Freudenberg et al. (1989) who suggested that ELISA determinations of relatively

adhesive antibodies, such as IgM, to amphiphilic molecules, such as lipid A, can often

lead to false positive results interpreted as cross-reactivity. It has also been shown

that LOS from pathogenic Neisseria and Haemophilus species contain terminal

tetrasaccharides that are homologous to paragloboside (Preston et al., 1996). Thus, it

is feasible that LOS of oral bacteria could be recognised similarly by the host, and

present shared epitopes as a strategy to evade the immune response.

With certain periodontopathogens, the detectable antibody response to these types of

antigens may provide some useful diagnostic information. The validity of targeting

LPS from oral pathogens as an immunodominant antigen is still fraught with

unknowns. Studies need to be carried out in the future to identify if naturally

protective responses to LPS exist. If they do, the information may be of importance in

prevention or intervention through vaccine implementation.

1.6.5 Heat shock proteins

Heat shock proteins (HSP) or stress induced proteins are produced by both prokaryotic

and eukaryotic cells in response to various insults, such as a sudden increase in

temperature (Kaufmann, 1990). HSP function to protect the cell during these adverse

environmental changes. As well as being molecular chaperones (Georgopoulos,

92

1992), it has recently been suggested that they are important in host immune responses

and the pathogenesis of disease (Panchanathan et al., 1998).

Immunoreactivity of HSP was initially recognised by the homology between a highly

immunoreactive 65 kDa protein of Mycobacterium and a 60 kDa HSP of E. coli

(Shinnick et al., 1988; Thole et al., 1988). It was suggested that this protein was also

related to a strongly immunoreactive antigen found in virtually all Gram negative

bacteria. Immunoreactive proteins in many prokaryotes have been identified which

show homology to HSP, especially the HSP90, HSP70, HSP60 and low molecular

weight families (Shinnick et al., 1991). The HSP70 family is highly conserved, with

the amino acid sequence of the human HSP70 having 50% identity with the E. coli

and 73% with the Drosophila HSP70 family. There have been reports demonstrating

that some bacterial HSP are immunogenic proteins (Buchmeier et al., 1990; Wu et al.,

1994; Perez-Perez, 1996).

Tang et al. (1997) studied the immunodominant nature of Salmonella typhi HSP60

(GroEL). This report showed that HSP of pathogens are "immunodominant" antigens

and targets of the host immune response. These results were in agreement with other

studies and were confirmed by substantial production of antibody to these antigens.

Further studies indicating the importance, and purported immunodominance of HSP

include investigations of GroEL and GroES of Camphylobacter jejuni (Wu et al.,

1994) and Mycobacterium leprae (Mehra et al., 1992). These reports supporting an

important role for HSP and host responses to these antigens in other infectious

diseases, naturally lead to an evaluation of the potential role of these macromolecules

in periodontitis.

A recent study by Ando et al. (1995) examined the production of HSP by the

periodontal pathogens: A. actinomycetemcomitans; Eikenella corrodens;

Fusobacterium nucleatum; P. intermedia; Prevotella nigrescens; Prevotella

melaninogenica and Treponema socranskii. All of the HSP produced by these

bacteria reacted with anti-Yersina enterocolitica HSP60 antibodies and/or monoclonal

antibodies to the 65 kDa HSP of Mycobacterium. Additionally, these authors

investigated soluble HSP in inflamed gingival tissue of chronic periodontitis patients.

93

Gingival homogenate samples from 3 of 4 patients reacted with anti-human HSP60

and anti-bovine brain HSP70. However, none of the homogenate samples reacted

with antibodies against bacterial HSP.

Maeda et al. (1994), cloned a HSP60 (GroEL) homologue from P. gingivalis using the

E. coli GroEL gene as a probe. The sequence of this P. gingivalis GroEL HSP

showed about 59% amino acid identity with E. coli GroEL and the 65 kDa

Mycohacterium tuberculosis antigen, as well as with the human HSP60 sequence. An

immunogen immunoblot analysis, using purified P. gingivalis GroEL, revealed that it

was recognised by antibody in 8/10 sera from periodontitis patients, and in 3/9 healthy

controls. These findings are consistent with the association of P. gingivalis with

periodontitis. In addition these antibodies are present in response to a low level,

continuous or transient exposure of the healthy subjects to P. gingivalis GroEL and

GroEL homologues from many other micro-organisms, or even host molecules.

Hinode et al. (1998), attempted to address the cross-reactive potential of HSP. They

evaluated the N-terminal amino acid sequences of GroEL (HSP60 family) and DnaK

(HSP70 family) like proteins from P. gingivalis, A. actinomycetemcomitans and B.

forsythus, in relation to potential cross-reactivity of antibodies. Antibodies specific

for DnaK-like proteins appeared to be primarily induced against non-conserved

epitopes of the protein. The DnaK-like proteins from P. gingivalis and B. forsythus

reacted only very weakly with antibodies specific for the DnaK of E. coli. However, a

clear cross-reactivity between antibodies raised to GroEL-like proteins of P.

gingivalis, A. actinomycetemcoin itans and E. coli was detected. This indicated that a

significant portion of the antibodies induced against GroEL-like proteins are directed

at epitopes conserved across bacterial genera. Why antibodies are produced in

response to conserved epitopes of one HSP and to primarily non-conserved epitopes

of another HSP is unknown. This could be a strategy adopted by the pathogens to

increase their virulence. It may be a subversion of host responses by alteration of the

efficiency of antigen processing and presentation by the host immune system.

A summary of the literature with respect to the role of HSP in the host response of the

periodontium suggests two main opinions; One view proposed by Ando et al. (1995)

94

is that periodontopathic bacterial HSP antigens form immune complexes with local

antibodies, which activate or exacerbate the inflammatory response, enhancing tissue

damage. However, the lack of detection of bacterial HSP in gingival homogenate

samples, and evidence of host derived HSP emphasises the potential for chronic

inflammation to stress local tissues during destructive periodontitis (Kaufmann,

1990). The second view is that antigenic cross-reactivity poses a potential problem in

evaluating HSP as immunodominant antigens of these pathogens. The term "common

antigens" is indicative of their highly conserved nature. As discussed previously with

LPS it may be difficult to ascertain if the detection of a host immune response

documents: (i) the induction of elevated antibodies to a particular HSP of the oral

bacterial pathogen under investigation; or (ii) cross-reacting antibodies elicited by

other commensal or pathogenic bacteria, or even autoantibodies to human HSP.

Bangsborg et al. (1989) reported the problems in documenting dominant serological

reactions using HSP, as they induce antibodies that are strongly cross-reactive.

However, antibodies against HSP have been implicated as "immunodominant" in

various autoimmune diseases including Systemic Lupus Erythematosus (Jarjour et al.,

1991) and rheumatoid arthritis (Van Eden, 1990).

This is clearly an exciting area of investigation in the microbial pathogenesis of

periodontal disease. Nevertheless, creative experimental designs and approaches will

be required to identify the functional importance of these immune responses. In

addition, the significance of HSP to oral bacterial virulence strategies, and the

justification for considering these immunogens in vaccine development will need to

be evaluated.

1.6.6 Leukotoxin

A number of Gram negative bacteria produce an array of diffusable protein molecules

which belong to the repeats-in-toxin (RTX) family of bacterial exotoxins. These

exotoxins are cytolytic proteins which are generally heat labile, induce tissue necrosis

and can be used to stimulate protective immune responses (e. g. tetanus toxoid)

(Nisengard and Newman, 1994). Members of the RTX exotoxin family are involved

in a variety of different bacterial infections and target different cell types and tissues

95

specific for the associated disease entities. There are a number of features that are

possessed by all members of this family, including: (i) the genetic organisation; (ii)

certain structural features; and (iii) functional features such as haemolytic, leukotoxic

and leukocyte-stimulating activities.

Of the oral micro-organisms associated with periodontitis, A. actinonrycetemcomitans

produces a leukotoxin which is a member of the RTX family. This is heat labile,

soluble and was initially reported because of its ability to destroy PMN (Baehni et al.,

1979; Baehni et al., 1980). However, recent studies indicate that it activates apoptotic

processes in a variety of cell types (Mangan et al., 1991; Kato et al., 1995; Korostoff

et al., 1998). At high concentrations, leukotoxin appears able to bind to a PMN target

cell membrane, to form pores in the membrane and to kill the cell through osmotic

lysis. This releases intracellular lysosomal products and exacerbates tissue damage.

Initial studies of antibody responses to A. actinomycetemcomitans sonicate antigens

and leukotoxin in humans suggested that the characteristics of the responses generally

paralleled each other (Ebersole, 1990). Importantly, patients with A.

actinomycetemcomitans infections generally demonstrated extreme elevations in

antibody to this antigen (Ebersole et al., 1991 a; Califano et al., 1989; Gunsolley et al.,

1991), suggesting that it was a strong immunogen. This has been confirmed in rodent

models (Shenker et al., 1993). The human response showed elevated leukotoxin

antibody consistently in localised aggressive periodontitis patients (McArthur et al.,

1981; Tsai et al., 1981; Ebersole et al., 1983). With regard to the issue of

immunodominance, a crucial point is that the antibody to the leukotoxin was not a

dominant antibody response in these patients. As mentioned previously, the main

response was to LPS and serotype determinants of A. actinomycetemcomitans. It was

concluded that LPS is an immunodominant antigen. Therefore, the response to this

molecule, should be the focus of efforts to immunologically interfere with the

pathogenicity of A. actinomycetemcomitans. The biologic relevance of leukotoxin as

a virulence factor has not been clearly demonstrated in vivo. Nevertheless, recent

genotypic studies support the concept that A. actinomycetemcomitans isolates which

have an altered promoter region, resulting in enhanced leukotoxin production, are

most frequently associated with disease sites in selected populations (DiRenzo et al.,

1994; Haubek et al., 1995). Recent data suggests that leukotoxin antibody levels may

96

be more discriminatory than antibody to the whole bacteria (Califano et al., 1997).

Furthermore, one report has suggested that leukotoxin challenge of a human immune

system reconstituted into mice, induced antibodies which correlated with some

protection from lesion formation (Shenker et al., 1993). To conclude, the reactivity of

humans to leukotoxin cannot be considered a `dominant' response. However, there

appear to be `dominant epitopes' on this antigen which can consistently elicit antibody

in infected subjects which appears to ameliorate the effects of this toxin.

Consequently, the use of antibody to leukotoxin as a biomarker of an immunological

target in periodontal disease seems worthy of continued investigation.

1.6.7 Cysteine proteases

The cysteine proteases of P. gingivalis clearly contribute to the physiological function

of this asaccharolytic bacterium. They are of great importance in the nutrition and

growth of the micro-organism (Kesavalu et al., 1997). They also appear to play a

crucial role in the survival of the bacteria through proteolytic elimination of opsonins

at the bacterial cell surface (Cutler et al., 1993; Schenkein et al., 1995), via

degradation of host response communication molecules including cytokines and

chemokines (Steffen et al., 1999). These enzymes also appear to correlate with

pathogenicity of the bacteria (Curtis et al., 1993; Feuille et al., 1996; Kesavalu et al.,

1998). They aid the adherence of P. gingivalis to erythrocytes (Nishikata et al., 1991;

Curtis et al., 1993; Pike et al., 1996), neutrophils (Lala et al., 1994), fibronectin

(Lantz et al., 1991a), fibrinogen (Lantz et al., 1991b), collagenous substrata (Naito et

al., 1988), as well as to other bacteria (Ellen et al., 1992). These proteases can also

directly contribute to the induction and alteration of the inflammatory response

(Potempa and Travis, 1996).

Curtis et al. (1996) studied 3 types of cysteine proteases associated with P. gingivalis,

each demonstrating a specificity for arginine containing peptide bonds (RgpA-1,

RgpA-1A and RgpA-IB). Similar reports have provided evidence of additional

cysteine proteases with varied specificity, which are produced by this pathogen (Bedi,

1994; Chen et al., 1992; Pavloff et al., 1995). In order to illustrate the relationship

97

between these proteases and the host response to P. gingivalis, there follows a review

of the antigenic features RgpA-1.

RgpA-1 or HRgpA is a heterodimer of approximately 50 kDa, which is composed of

an a and ß segment or chain. RgpA-1 A is a monomer of a free a segment and RgpA-

lB appears to be a post-transcriptionally modified form of the a segment (Curtis et

al., 1996). One study by Curtis et al. (1993), used polyvalent rabbit antiserum raised

against RgpA-IA. They demonstrated a recognition of all 3 protease types, indicating

that the a chains are immunochemically related. However, 4 of 5 representatives

from a panel of monoclonal antibodies recognised only RgpA-1. This indicated

specificity for the (3 segment portion of this protease, which is the non-catalytic part of

the enzyme. The a segment bears the protease active site. The ß segment has a

primary sequence similar to that of adhesins and haemagglutinins of other molecules

and micro-organisms. Why the immune response is directed towards the seemingly

non-destructive portion of the enzyme remains unknown. The ß segment appears

important in adherence. Therefore, the host may target antibodies towards this portion

to block attachment and access of the protease to the host cells or proteins.

Alternatively, the ß segment may have other unknown functions which also increase

its immunogenicity. In contrast, this may be a deliberate ploy of the bacteria to

increase its virulence potential by diverting antibody responses to a less-relevant

portion of the enzyme.

Nevertheless, these proteases, are implicated in periodontal disease, and induce the

production of specific IgG in humans. As mentioned above, P. gingivalis appears to

be effective in neutralising humoral elements of host defence. It uses trypsin-like

proteases which degrade all isotypes of immunoglobulins (Gregory et al., 1992) and

many complement factors (Schenkein, 1988). Additionally, these proteases have been

shown to give P. gingivalis the ability to degrade many human proteins (Kilian, 1981).

Thus, clarifying the relationship between the systemic response to this antigen(s) and

disease progression would appear useful, since this appears to be an important

virulence factor enabling this bacteria to avoid host immune defences. Moreover,

other reports have demonstrated the effective binding of IgG to virulent bacteria such

98

as P. gingivalis, has a crucial role in complement activation and also in the

opsonisation and phagocytosis of this organism (Cutler et al., 1991a, b). Increased

antibody levels to a trypsin-like protease produced by P. gingivalis was identified in

early studies of humoral immune responses in periodontitis patients (Ismaiel et al.,

1988). Aduse-Opoku et al. (1995), recently characterised a cell surface or extra-

cellular arginine-specific protease antigen (i. e. related to the trypsin-like protease

activity). This group also characterised this protease as a functionally important target

of the host immune response (Curtis et al., 1996). It has recently been demonstrated

that active immunisation of non-human primates with a cysteine protease (e. g.

porphypain-2) elicited significant antibody responses in the animals (Mortiz et al.,

1998). It also altered the microbial ecology in disease states, and significantly

affected the progression of bone loss in the animals. Booth and Lehner (1997) have

also produced a murine monoclonal antibody to P. gingivalis which retarded re-

colonisation of deep pockets by this pathogen in periodontitis patients. Interestingly,

this antibody inhibited the hemagglutination of red blood cells, which is related to

certain functions of the cysteine proteases of P. gingivalis.

Importantly these relatively few human studies have generally shown an increased

prevalence of antibody to these proteases in periodontitis patients. However, the

antibody levels were clearly not dominant in the response to this pathogen. This

finding highlights the potential uncertainty in the field when identifying critical

antigens relative to antibody levels and functional capacity to protect the host.

However, this aspect of the host-parasite interaction will remain a high profile

research area with the goal of evaluating pathogenic mechanisms of P. gingivalis, as

well as a potential immunologic targets.

1.6.8 Superantigens

Increased knowledge of the molecular mechanisms for cell co-operation and the cell

surface molecules required for adaptive immune induction, has revealed that not all

antigens which bind to the MHC class II complex are presented as peptides in the

peptide binding groove. Hence, they do not result in the development of a specific

immune response by the conventional method. One such class of antigens, identified

99

within various bacterial and viral protein repertoires, has been termed superantigens

(Janeway and Travers, 1994).

Superantigens are not processed by antigen presenting cells and are not presented by

MHC molecules in the conventional manner. Instead of binding to the peptide

binding groove of the MHC molecule, superantigens bind directly to the lateral

surface of both the MHC class II molecule and the Vß region of the TCR. The

superantigen is recognised solely by the V region of the ß chain of the TCR (Kappier

et al., 1989). In contrast, to conventional antigen processing and presentation, the a

chain V region and the DJ junction of the ß chain have little or no effect on

superantigen recognition. This recognition process gives superantigens the capability

of stimulating a large number of T cells (4-5% of the T cell population) in a non-

specific manner (Kappler et al., 1989). This reaction with a large number of T cells

gives rise to the characteristics associated with superantigens: (i) pyrogenic effects in

excess of those produced by LPS; (ii) inhibition of immunoglobulin synthesis; and

(iii) interference with LPS clearing by the liver causing a massive increase in the

lethality of bacterial LPS (Daniel and Van Dyke, 1996).

Superantigens are known to be produced by a number of pathogenic micro-organisms,

including some found in the oral cavity. However, to date no clear role for

superantigens in periodontal disease has been demonstrated. Mathur et al. (1995)

investigated superantigen like activities of putative periodontopathogens in both

periodontally healthy and diseased subjects. Co-culturing peripheral blood

mononuclear cells with P. intermedia resulted in an increased expression of V(32+,

V135+ and V06+ cells in both healthy and diseased subjects. However, similar

treatment with either P. gingivalis or A. actinomycetemcomitans showed no changes

in the Vß TCR repertoire. Studies by Nakajima et al. (1996) and Zadeh and Kreutzer

(1996) which involved profiling gene expression in gingival tissues, suggested that

expression of the Vß family in tissue borne T cells was not random. However,

limiting an explanation for this finding to superantigen processes does not account for

a skewing towards particular Vß receptor families. This could be due to the presence

of bacteria in the subgingival plaque biofilm selectively stimulating cells of a

100

particular V(3 receptor family (Mathur et al., 1995). Alternatively, homing of cells to

the diseased tissue may be mediated, in part, by Vß expression (Taubman et al.,

1994). In addition, retention of specific T cells in the tissues may be due to the

characteristics of their VP receptors and the presence of specific antigens. Therefore,

our knowledge reflects a number of ill-defined and poorly understood processes,

especially as related to the existence and importance of superantigens in periodontal

disease. A more detailed evaluation of the available studies provides equally

conflicting results. Vß8 cells are over expressed (Zadeh and Kreutzer, 1996) or under

expressed (Nakajima et al., 1996). Alternatively, Petit and Stashenko (1996)

demonstrated that extracts of periodontopathic bacteria do not stimulate T cells at all

in a superantigenic manner when tested in a murine model. It is not clear if these

conflicting reports are due to different types of superantigens, different sampling

strategies, or even just variation in patient populations. The disparate findings need to

be clarified in order to define the utility of the Vß chains as markers of superantigen

activity within the host-parasite interactions, in periodontal disease.

However, one might reconsider whether there is an a priori case for emphasising the

contribution of superantigens in periodontal disease. This disease has been described

as a chronic infection involving an ongoing humoral immune response, shown by the

presence of predominantly antibody secreting cells in the periodontal lesion (Kinane

& Lindhe, 1997). In contrast, infections in which the pathogens present superantigens

as part of their virulence strategy, are generally acute, with a rapid onset of symptoms,

and immunoglobulin production is usually suppressed. This profile is not observed in

periodontitis and systemic pyrogenic activity (another characteristic of infections with

superantigen possessing micro-organisms) is an exceptional occurrence in

periodontitis patients. Therefore, at present it is questionable whether a role for

superantigens is consistent with the characteristics of this disease.

1.6.9 Other strategies for identifying immunodominant antigens

A recent study (Kawai et al., 1998) attempted to identify an immunodominant antigen

in chronic periodontitis using monoclonal antibodies. Mouse monoclonal antibodies

101

were raised against crude antigen preparations of P. gingivalis. The antigenic

specificity of the monoclonal antibodies were then compared with those of serum

antibodies detected in chronic periodontitis patients and healthy subjects. Binding of

one monoclonal antibody was inhibited by serum antibody molecules present in all

chronic periodontitis patients. This serum antibody was absent in the polyclonal

antibody population in the control subjects however. The inference was that the P.

gingivalis antigenic epitope identified by this murine monoclonal antibody could

represent an important immunogen in humans infected with this pathogen. The

antigen recognised by this monoclonal antibody was demonstrated on the cell surface

of P. gingivalis, and the epitope was composed of carbohydrate residues independent

of the LPS antigen. Thus, new technological and conceptual advances will continue

to challenge existing paradigms of immunodominant antigens in periodontal disease,

and provide additional tools for addressing this important concept.

As mentioned previously, the aim of this text is to provide an overview of the

approaches and strategies which have been used in the past and those which are

currently being employed in periodontal infections, to examine the concept of

immunodominant antigens. It appears there is no clear way ahead for the concept of

immunodominance in periodontal disease research. This polymicrobial infection is

complex due to the intertwining of the relationships within the microbial ecology.

The plethora of host and microbial variables may make successful exploitation of any

interventive strategy, based on the immunodominance concept, difficult to implement

in this disease. This concern is exemplified by the breadth of studies supplying

disparate findings regarding the antigenic specificity of dominant host responses to the

putative pathogens. The identification of immunodominant antigens may be relevant

if additional supportive knowledge is gained in the future. In this case it is suggested

that utilisation of these critical antigens for screening and diagnosis of the disease, as

well as using them as targets for vaccine development, will not only be feasible, but

desirable.

102

1.7 Pathogenesis of Periodontal Disease

1.7.1 Introduction

The normal pink in colour and firm in consistency healthy gingiva should be free from

histological evidence of inflammation (figure 1.2). But this is rarely the case, and

however healthy an individuals gingiva may be, when seen through the microscope

they are always slightly inflamed due to the ever present plaque bacteria (figure 1.3).

The inflammatory cells which have infiltrated the healthy gingiva are predominantly

PMN that are associated with the junctional epithelium, and lymphocytes that are

found in the subjacent connective tissue (Page and Schroeder, 1976). All the

characteristics of the inflammatory response, as previously described, are seen in the

gingiva even in health. These include; increased vascular permeability and

accumulation of fluid, plasma proteins and leukocytic cells into the gingival crevice

region, to create GCF. The reason for the migration of cells into the junctional

epithelium as well as the crevice is due to the production of cytokines and various

other products by the host and bacteria. This subject is discussed in more detail in

section 1.4.

A further indication that inflammation is present in a `healthy' gingiva is that

upregulation of the adhesion molecules ICAM-1 and ELAM-1 have been shown in

health (Moughal et al., 1992; Kinane et al., 1991).

Although, as discussed above, an inflammatory response is thought to occur even in

health, a balance has to be maintained for the gingiva to remain `healthy'. Although

bacterial plaque is present and the host responds to it by initiating an inflammatory

response, the micro-organisms and the host are probably in equilibrium with each

other. However, the gingiva stops being classified as `healthy', and gingivitis follows

when the micro-organisms in the plaque initiate a substantial inflammatory response.

103

1.7.2 Gingivitis

Gingivitis can occur after 10-20 days of plaque accumulation. It is detected due to

redness and swelling of the tissues which have an increased tendency to bleed on

probing. Histologically, there has been a massive influx of fluid and cells into the

tissues. Dilation of arterioles, capillaries and venules increases hydrostatic pressure

within the microcirculation leading to increased intercellular gaps between adjacent

capillary and endothelial cells. The cells that migrate into the tissues are mainly

lymphocytes, macrophages and PMN. Experimental work, particularly studies

involving dogs (Lindhe and Rylander, 1975) have shown that when there is a cellular

infiltrate, lymphocytes are found to adhere to the collagen matrix as they are possibly

primed to periodontally relevant antigens, and therefore, remain in the tissues (Kinane

and Lindhe, 1997). The PMN which infiltrate the tissues tend to migrate into the

gingival sulcus and do not remain in the tissues.

In addition to the influx of inflammatory cells into the tissues, there is increased flow

of plasma proteins, antibodies, complement proteins and protease inhibitors into the

area (Payne et al., 1975; Schroeder et al., 1975). The flow of GCF also increases, so

that toxins and substances released from microbes are washed out of the tissues and

the gingival crevice. As the inflammatory response to the plaque bacteria increases

there is also a marked expression of adhesion molecules to assist the migration of

leukocytes out of the blood vessels, and into the tissues and crevice (Kinane and

Lindhe, 1997).

After approximately 7 days the lesion is referred to as the early gingival lesion (Page

and Schroeder, 1976) (figure 1.4). In the early gingival lesion, lymphocytes and PMN

are the prominent infiltrating cells and very few plasma cells are noted at this time

(Listgarten and Ellegaard, 1973; Payne et al., 1975; Seymour et al., 1983; Brecx et al.,

1987). At this stage the inflammatory cell infiltrate comprises approximately 15% of

the connective tissue volume. However, due to fibroblast degeneration and collagen

destruction that occurs in the tissue, further inflammatory cell infiltration can occur as

space is created.

104

It is not known how long the gingivitis lesion remains like this or when it returns to

health or progresses to an increased inflammatory state. There appears to be a

significant variation between individuals however, probably based on individual

predispositions and susceptibility.

If the plaque remains present, the inflammatory state continues and becomes

exacerbated. Continuation of the inflammatory response sees the influx of further

exudative fluid, inflammatory cells and plasma proteins. The tissues become even

more swollen and oedematous (Löe and Holm-Pedersen, 1965), however, at this stage

there is no bone loss nor apical epithelial migration evident. According to the

classification determined by Page and Schroeder (1976) the lesion is termed the

established lesion (figure 1.5). The main difference between the early and established

lesion is not only the progression of the inflammatory immune response, but the

established lesion contains many more mature plasma cells. Prominent infiltration of

PMN is also observed in the junctional and sulcular epithelium (Brecx et al., 1988;

Lindhe et al., 1980; Okada et al., 1983). The latter may proliferate and migrate into

the infiltrated connective tissue, along the root surface and extend deeper into the

connective tissue. The gingival sulcus deepens and the coronal portion of the

junctional epithelium is converted into `pocket lining' epithelium. This is not

attached to the tooth surface and has a heavy leukocyte infiltrate, predominantly made

up of neutrophils which eventually migrate across the epithelium into the gingival

crevice or pocket.

It is thought that two different types of established lesion exist. One is thought to be a

more stable lesion that can remain this way for long periods of time, and the second is

much more aggressive and converts to progressive destructive lesions (Lindhe et al.,

1975; Page et al., 1975).

1.7.3 Periodontitis

The advanced gingival/periodontal lesion is the final stage of the disease. The

advanced lesion differs from the established lesion by alveolar bone loss being

105

evident. There is also evidence of extensive fibre damage and apical migration of the

junctional epithelium (Lindhe et al., 1980; Listgarten and Hellden, 1978; Seymour and

Greenspan, 1979; Seymour et al., 1979). The inflammatory immune response

continues to progress, and the inflammatory cell infiltrate extends laterally and

apically into the connective tissue. Epithelium layers are broken down, re-growing at

a more apical location. Pocket formation occurs, bone is resorbed and granulation

tissue forms, which is heavily vascularised and full of antibody producing plasma

cells (Kinane and Lindhe, 1997). Plasma cells are the predominant cell type in the

advanced lesion (Garant and Mulvihill, 1972) (figure 1.6).

The progression of the inflammatory response continues in the presence of plaque

bacteria and the microbes that have continued to inhabit the niches that have been

created through tissue damage and the frustrated inflammatory response. When no

oral hygiene improvements are made or treatment pursued the situation is worsened.

Microbes will continue to produce noxious substances, the host inflammatory

response will be perpetuated and tissue damage will continue. This is seen by

continuous deepening of pockets, extending granulation tissue lesions, continuous

bone and ligament loss and eventually loss of the teeth (Kinane and Lindhe, 1997).

The progression from gingivitis to periodontitis varies in time between individuals.

Brecx et al. (1988) suggested that at least 6 months is required for gingivitis to

progress to periodontitis in the normal situation. The reason some individuals go on

to develop periodontitis and others do not and the length of time this takes in different

patients is thought to be multifactorial. Seymour et al. (1979) suggested that it is a

shift from B cell to T cell dominance that causes the transition from gingivitis to

periodontitis. This theory has been contested by a number of authors (Gillett et al.,

1986; Page, 1986) and it is generally accepted that an advanced lesion is

predominantly a plasma cell lesion (Garant and Mulvihill, 1972; Kinane and Lindhe,

1997).

106

Figure 1.2 The pristine gingiva

Figure 1.2 shows a pristine gingiva which is almost free from histological evidence of inflammation. This is rarely seen in health.

Figure 1.3 The normal healthy gingiva

Figure 1.3 shows a normal healthy gingiva. This picture of microbial colonisation

due to ever present plaque bacteria is usually seen in health. Inflammatory cells have

begun to infiltrate the junctional epithelium and connective tissue. These are mainly

PMNs, monocytes, macrophages and lymphocytes.

Figure 1.4 The early gingival lesion

Figure 1.4 shows the early gingival lesion. This is thought to occur after

approximately 7 days. The number of infiltrating inflammatory cells has increased

and comprises about 15% of the connective tissue volume. The infiltrate is mainly

PMNs, macrophages, monocytes and lymphocytes. Very few plasma cells are seen.

Figure 1.5 The established gingival lesion

Figure 1.5 shows the established gingival lesion. The inflammatory immune response

has progressed, and the connective tissue is comprised of 10-30% plasma cells.

Clinically marked proliferation of the junctional epithelium is seen, however, bone

loss and apical epithelial migration are not evident.

Figure 1.6 The advanced lesion

Figure 1.6 shows the advanced lesion seen in periodontitis. The inflammatory cell

infiltrate has progressed laterally and apically and is comprised of predominantly

plasma cells. Clinically there is evidence of extensive fibre damage, apical migration

of the junctional epithelium and bone loss. Figures adapted from Kinane and Lindhe

(1997).

1.8 Risk Factors

1.8.1 Introduction

The term "risk factor" can be defined as "an aspect of personal behaviour or lifestyle,

an environmental exposure, or an inborn or inherited characteristic, which on the basis

of epidemiological evidence is known to be associated with a health related condition"

(Last, 1988). Therefore, the presence of a risk factor implies that there is an increased

chance of the disease occurring.

Periodontal disease is not a consequence of pathogens alone, and these are not

sufficient on their own to cause the disease (Socransky and Haffajee, 1992). The

presence of micro-organisms is a crucial factor, but the progression of the disease is

related to host based risk factors.

1.8.2 Age

There can be little doubt that the prevalence of periodontal disease increases with age

(Salvi et al., 1997). However, it is not clear if age leads to increased susceptibility or

that there are cumulative factors which control the onset of periodontitis. Homing et

al. (1992), reported that age is a risk factor for periodontal disease. However, other

authors (Holm-Pederson et al., 1975; Machtei et al., 1994) have indicated that this is

not the case before the age of 70 years, and until this age the rate of periodontal

destruction is the same throughout adulthood.

1.8.3 Low Socio-economic status

Low socio-economic status is generally associated with more severe periodontitis and

data from the U. S. public health service (1965,1979) indicates this. However, in

studies where periodontal status is adjusted for oral hygiene and smoking, the

associations between lower socio-economic status and more severe periodontal

disease are not observed (Grossi et al., 1994; Grossi et al., 1995).

110

1.8.4 Smoking

There has been a substantial amount of literature reporting the detrimental effect of

smoking on periodontal health. There is a consistent indication that smoking is a risk

factor for periodontal disease and there is a positive correlation between attachment

loss and smoking (Arno et al., 1959; Ah et al., 1994; Bergstrom, 1989; Haber, 1994;

Haber and Kent, 1992; Haber et al., 1993; Horning et al., 1992; Preber and Berström,

1990). In many of the studies on this subject, multivariate analyses are carried out to

ensure that smoking is an independent risk factor for periodontal disease, and that any

results showing a correlation are not due to poor oral hygiene, plaque, socio-economic

status or any other risk factor. The risk for periodontitis is considerable if an

individual uses tobacco products, with estimated ratios being of the order 2.5 to 7.0 or

even greater for smokers compared with non-smokers (Salvi et al., 1997).

The mechanisms by which smoking affects periodontal status is not very well

understood (Haber, 1994). Studies have indicated that the occurrence, relative

frequency, or combinations of micro-organisms associated with destruction do not

differ between smokers and non-smokers (Preber et al., 1992). However, perhaps

smoking has a direct effect on the inflammatory and immune mechanisms. On this

theme, smokers with periodontal disease have less clinical inflammation (Feldman et

al., 1983) and gingival bleeding (Bergstrom and Floderus-Myehed, 1983) compared

with non-smokers. The biological effect on the inflammatory immune response due to

smoking could be as a result of the by-products of tobacco such as nicotine, which

causes local vasoconstriction reducing blood flow, oedema and clinical signs of

inflammation.

1.8.5 Systemic disease

There is strong evidence that periodontal disease enhances the risk for many systemic

diseases, including atherosclerosis, heart disease, stroke and the births of infants with

low birth weight (Beck et al., 1996; Offenbacher et al., 1996). There is also evidence

that certain systemic diseases can adversely effect host defence systems and, therefore,

act as a risk factor for gingivitis and periodontitis.

111

Neutropenic states have been associated with rapidly progressive forms of periodontal

disease (Miller et al., 1984). Radiographs taken in some cases have indicated

involvement of the periodontal tissues, showing bone loss around the primary and

secondary teeth (Miller et al., 1984).

There is an association between the disease leukocyte adhesion deficiency (LAD) and

periodontal disease. Patients suffering from this disease have an abnormal gene that

codes for the family of adhesion molecules called the leukocyte integrins. Patients

with this deficiency have a dramatically increased susceptibility to infections and

suffer from a very severe form of periodontal disease that can lead to premature loss

of deciduous and permanent teeth (Crawford, 1994).

Diabetes Mellitus has been linked to an increased risk of periodontal disease (Bacic et

al., 1988; Bartolucci and Parkes, 1981; Belting et al., 1964; Grossi et al., 1994;

Hugoson et al., 1989). Type I diabetes mellitus is also known as insulin dependent

diabetes mellitus, and commonly develops before 30 years of age. Type II, or non-

insulin dependent diabetes mellitus is usually found to develop after the age of 40

years. A strong correlation has been found between aggressive periodontitis and type

I diabetes mellitus, especially in patients with raised blood glucose levels (Bissada et

al., 1982; Cutler et al., 1991; Manouchehr-Pour et al., 1981a; Manouchehr-Pour et al.,

1981b). Type II diabetes has also been linked with periodontal disease (Beck et al.,

1990; Grossi et al., 1994). However, differences in oral health between type I and

type II diabetes occur and may be related to differences in glycaemic control, age,

duration of disease or periodontal susceptibility (Salvi et al., 1997). Well-controlled

diabetics are more likely to be like non-diabetics in their periodontal status, and those

with poorly controlled diabetes are likely to be more susceptible to disease (Mealey,

1996).

1.8.6 Other possible risk factors

Preliminary studies have indicated an association between stress, distress and coping

behaviour (Moss et al., 1996). The rationale behind many of the studies investigating

112

a link between periodontal disease and the psychoemotional state are that these factors

can lead to depression of immune responsiveness to putative periodontopathogens.

1.8.7 Genetics

Many studies have been reported over the past few years regarding a possible genetic

basis for periodontal disease, and many of these studies have been reviewed (Hart,

1994; Hart, 1996; Hart and Komman, 1997; Michalowicz, 1994; Aldred and Bartold,

1998). The generally accepted dogma at present is that the aggressive forms of

periodontitis have a heritable pattern with a single gene disorder (Aldred and Bartold,

1998). With regards to chronic periodontitis, there is a split in opinion, with a general

surmise being that there is little genetic influence in the clinical manifestation of this

disease opposing recent evidence that suspects a genetic susceptibility associated with

the disease (Aldred and Bartold, 1998).

The exact genetic link with both aggressive and chronic forms of periodontal disease

have yet to be identified. More work has been carried out investigating the link with

aggressive periodontal disease, and although many studies have been undertaken the

conjectures made from these have not been fully accepted. Specific chromosomal

anomalies have been investigated, and the outcomes appear to indicate that even with

a small disease entity such as localised aggressive periodontal disease, there is

probably considerable genotype heterogeneity even if the clinical manifestations

appear to be very similar. This can be observed by the fact that different groups

carrying out genetic linkage studies have found such varying results (Boughman et al.,

1986; Roulsten et al., 1985; Hart et al., 1993; Wang et al., 1996).

Studies have also been carried out to investigate a link between periodontal disease

and genetic markers of a more general nature. Again, most reports made are

controversial on this issue. A number of studies have associated HLA antigens with

generalised aggressive periodontitis (Katz et al., 1987; Reinholdt et al., 1977; Shapira

et al., 1994), however, others have found no association at all (Cullinan et al., 1980;

Saxen and Koskimies, 1984).

113

Other genes have been suggested to be linked with periodontal disease including; the

gene responsible for IgG2 production (Marazita et al., 1994; Marazita et al., 1996),

FcyRIIa receptor (Wilson and Kalmar, 1996), prostaglandin endoperoxidase synthase

1 and other components of the cyclooxygenase pathway (Hart and Kornman, 1997)

and IL-1 (Kornman et al., 1997).

1.9 Aims and Objectives

1.9.1 Introduction

Many microbial pathogens have been proposed as playing a role in the induction of

the immune response in periodontal disease. The aim of the thesis was to investigate

the target of the humoral immune response in patients with chronic periodontitis.

This was achieved by a number of studies following a logical progression. The first

study investigated the systemic immune response by examination of the target of

serum antibodies in patients with chronic periodontitis. The project then developed to

investigate the target of the local immune response and to achieve this 3 different

techniques were utilised: hybridoma production; ELISPOT; and

immunohistochemistry.

1.9.2 Aims of Chapter 3-A Comparison of the changes in the humoral immune

response to antigens of periodontal pathogens following periodontal therapy

The aims of this part of the overall project were to:

1) investigate the effect of periodontal treatment in patients with chronic

periodontitis on polyvalent immunoglobulin, IgG and IgA titres against whole

fixed periodontal pathogens;

2) Investigate the effect of periodontal treatment on antibody titres to a large panel of

antigenic targets including whole fixed bacteria, outer-membrane protein

preparations and purified antigens;

3) Compare clinical and immunological parameters in chronic periodontitis patients

before and after periodontal treatment;

114

4) Investigate the phenomenon of cross-reactivity by examining cross-reacting

antigens to 4 periodontal pathogens.

1.9.3 Aims of Chapter 4- Immortalised B cells from periodontal disease patients

The aims of this part of the overall project were to examine the feasibility of creating

monoclonal antibodies from humans with periodontal disease by the:

1) Expansion of B cells numbers by utilisation of the CD40 system;

2) Fusion of human peripheral blood B lymphocytes and tissue plasma cells with a

suitable partner to obtain hybridomas;

3) Successful culture and cloning of antibody secreting hybridomas;

4) Screening of monoclonal antibodies specific for periodontal pathogens.

1.9.4 Aims of Chapter 5- Analysis of the target of antibody secreting cells from

diseased tissue of chronic periodontitis patients by ELISPOT

The aims of this section of the project were:

1) Examination of the target of antibody secreting cells isolated from tissue taken

from chronic periodontitis patients;

2) Comparison of the number of specific cells detected against 4 periodontal

pathogens.

1.9.5 Aims of Chapter 6- Detection of antibody-bearing plasma cells in periodontitis

granulation and gingival biopsy tissue

In this part of the thesis the aim was to:

1) Develop a technique to determine the specificity and numbers of plasma cells

present in tissue sections;

2) Examine the specificity of antibody bearing plasma cells in sections of diseased

tissue taken from patients with chronic periodontitis;

3) Compare the number and specificity of plasma cells in granulation and gingival

tissue samples detected using a panel of whole fixed periodontal pathogens,

sonicates and purified antigens.

115

1.9.6 Layout of the thesis

The following chapter covers the source of the equipment and chemicals used in the

various projects undertaken in this thesis. It also describes the basic methodology and

buffers used. Following chapter 2, the thesis has been divided into 4 further chapters

which describe individual studies: Comparison of the changes in the humoral immune

response to antigens of periodontal pathogens following periodontal therapy (chapter

3); Immortalised B cells from periodontal disease patients (chapter 4); Analysis of the

target of antibody secreting cells from diseased tissue of chronic periodontitis patients

by ELISPOT (chapter 5); and Detection of antibody-bearing plasma cells in

periodontal granulation and gingival biopsy tissue (chapter 6). In each of these

chapters the relevant introduction and literature review, methodology, results and

discussion of the results are presented. The organisation of this thesis is intended to

clarify the separate studies for the reader, as they are fundamentally different and if

grouped together could be confusing. These chapters are followed by a brief final

chapter (chapter 7) which summarises the findings of this thesis and discusses them in

the context of the aims of the overall project and the available literature on this

subject.

116

Chapter 2. Materials and Buffers

2.1 Introduction

The following chapter describes the materials and methodology for making the buffers

used routinely during the experiments presented in this thesis. Methods which are

specific to the individual chapters are found detailed in the methods section of the

relevant chapter.

2.2 Materials

The following list comprises the consumables used in the experiments presented in

this thesis and their sources.

2.2.1 Chapter 3 materials

PCP-12 Probe - Hu-Friedy Mfg Co., Chicago, U. S. A.

Florida Probe - Florida Probe Corporation, Florida, U. S. A.

Non-heparinised 7ml vacutainers - Becton Dickinson Vacutainer Systems Europe,

Meylan, France.

Fastidious anaerobe agar - LAB M, Bury, U. K.

Tryptic soy agar - Mast Diagnostics, Merseyside, U. K.

Horse blood -E&0 Laboratories, Bonnybridge, U. K.

N-acetylmuramic acid - Sigma Chemical Company Limited, Poole, U. K.

Columbia blood agar - Mast Diagnostics, Merseyside, U. K.

117

Sodium chloride (NaCl) - BDH Limited, Poole, England, U. K.

Potassium dihydrogen orthophosphate (KH2PO4) - BDH Limited, Poole, England,

U. K.

Di-sodium hydrogen orthophosphate 2-hydrate (Na2HP04.2H20) - BDH Limited,

Poole, England, U. K.

Potassium chloride (KCl) - BDH Limited, Poole, England, U. K.

Ethylenediaminetetracetic acid (EDTA) - BDH Limited, Poole, England, U. K.

Sodium hydroxide (NaOH) - BDH Limited, Poole, England, U. K.

Formal saline - BDH Limited, Poole, England, U. K.

Sodium azide (NaN3) - Sigma Chemical Company Limited, Poole, U. K.

Tris base - Sigma Chemical Company Limited, Poole, U. K.

Phenol red - Sigma Chemical Company Limited, Poole, U. K.

Sodium lauryl sarcosinate (SDS) - Sigma Chemical Company Limited, Poole, U. K.

BCA protein assay kit - Pierce & Warriner, Chester, U. K.

Immunlon 1B 96 well plates - Dynex Technologies, VA, USA.

Sodium carbonate (Na2CO3) - BDH Limited, Poole, England, U. K.

Sodium hydrogen carbonate (NaHCO3) - BDH Limited, Poole, England, U. K.

118

Hydrochloric acid (HC1) - BDH Limited, Poole, England, U. K.

Polyoxyethylene sorbitan monolaurate (Tween 20) - Sigma Chemical Company

Limited, Poole, U. K.

Lyophilised bovine serum albumin (BSA) - Sigma Chemical Company Limited,

Poole, U. K.

Skimmed milk powder - Marvel, Premier Beverages, Stafford, UK.

Biotinylated goat-anti-human immunoglobulin A- Sigma Chemical Company


Biotinylated goat-anti-human immunoglobulin G- Sigma Chemical Company


Biotinylated Goat-Anti-Human Polyvalent Immunoglobulin (IgA, IgG, IgM): Sigma

Chemical Company Limited, Poole, U. K.

Extravadin peroxidase - Sigma Chemical Company Limited, Poole, U. K.

TMB peroxidase substrate - Kirkegaard & Perry Laboratories, Maryland, U. S. A.

MRX II plate reader - Dynex Technologies, VA, U. S. A.

N-tris[Hydroxymethyl]methyl-2-aminoethanesulfonic acid (TES) - Sigma Chemical

Company Limited, Poole, U. K.

L-1-Chloro-3-[4-tosylamido]-7-amino-2-heptanone-HC1 (TLCK) - Sigma Chemical


119

Leupeptin - Sigma Chemical Company Limited, Poole, U. K.


Citrate treated 7ml vacutainers - Becton Dickinson Vacutainer Systems Europe,

Meylan, France.

L-cells - Schering-Plough, Dardilly, France

CRL-1668 cells - American Type Culture Collection, VA, U. S. A.

Foetal calf serum (FCS) - Gibco Life Technologies Limited, Paisley, U. K.

Penicillin - Gibco Life Technologies Limited, Paisley, U. K.

Streptomycin - Gibco Life Technologies Limited, Paisley, U. K.

L-Glutamine - Gibco Life Technologies Limited, Paisley, U. K.

Sodium pyruvate - Gibco Life Technologies Limited, Paisley, U. K.

RPMI 1640 medium - Gibco Life Technologies Limited, Paisley, U. K.

G-418 - Gibco Life Technologies Limited, Paisley, U. K.

Iscoves medium - Gibco Life Technologies Limited, Paisley, U. K.

Mitomycin-C - Sigma Chemical Company Limited, Poole, U. K.

Crystal violet concentrate - Pro lab Diagnostics, Merseyside, U. K.

Acetic acid - BDH Limited, Poole, England, U. K.

120

6 well plates - Gibco Life Technologies Limited, Paisley, U. K.

Histopaque-1077 - Sigma Chemical Company Limited, Poole, U. K.

Falcon tubes - Becton-Dickinson, Cowley, U. K.

Glacial acetic acid - BDH Limited, Poole, England, U. K.

Nigrosin - Sigma Chemical Company Limited, Poole, U. K.

Sodium azide - Sigma Chemical Company Limited, Poole, U. K.

Sheep red blood cells - Scottish Antibody Production Unit, Carluke, U. K.

2-aminoethylisothiouronium bromide (AET) - Sigma Chemical Company Limited,

Poole, U. K.

Sterile saline - BDH Limited, Poole, England, U. K.

Anti-human CD40 - Ancell Corporation, Bayport, MN, U. S. A.

Human interleukin-4 - ICN Biomedicals Limited, Oxon, U. K.

Dispase - Gibco Life Technologies Limited, Paisley, U. K.

Jockliks MEM medium - Gibco Life Technologies Limited, Paisley, U. K.

41µm nylon net filters - Millipore (U. K) Limited, Watford, U. K.

Neubauer haemocytometer - Sigma Chemical Company Limited, Poole, U. K.

121

Polyethylene glycol 1500 (PEG 1500) - ICN Biomedicals Limited, Oxon, U. K.

HAT concentrate - Sigma Chemical Company Limited, Poole, U. K.

Transferrin - Sigma Chemical Company Limited, Poole, U. K.

Insulin - Sigma Chemical Company Limited, Poole, U. K.

Ouabain - Sigma Chemical Company Limited, Poole, U. K.

HT concentrate - Sigma Chemical Company Limited, Poole, U. K.

Cell culture flasks (various sizes) - Gibco Life Technologies Limited, Paisley, U. K.


Multiscreen HA membrane-bottomed plates - Millipore Limited, Watford, U. K.

Tetanus toxoid - Evans Medical Ltd., Leatherhead, U. K.

Fungizone - Gibco Life Technologies Limited, Paisley, U. K.

HRP-conjugated avidin-D - ICN Biomedicals Limited, Oxon, U. K.

3-amino-9-ethylcarbazole (AEC) tablets - Sigma Chemical Company Limited, Poole,

U. K

N, N-Dimethylformamide - Sigma Chemical Company Limited, Poole, U. K.

Sodium acetate (CH3COONa. 3H20) - BDH Limited, Poole, England, U. K.

0.20µm filter - Millipore (U. K) Limited, Watford, U. K.

122


Paraffin wax - Bayer PLC., Newbury, U. K.

Silane-coated glass slides - BDH Limited, Poole, England, U. K.

Haematoxylin stain - BDH Limited, Poole, England, U. K.

Aluminium potassium sulphate (AIK(SO4)2) - BDH Limited, Poole, England, U. K.

Sodium iodate (NaIO3) - BDH Limited, Poole, England, U. K.

Citric acid (C(OH)(COOH)(CH2COOH)2. H20) - BDH Limited, Poole, England, U. K.

Choral hydrate (CC13CH(OH)2) - BDH Limited, Poole, England, U. K.

Calcium chloride (CaC12) - BDH Limited, Poole, England, U. K.

Eosin - BDH Limited, Poole, England, U. K.

Xylene - Genta Medical, Leeds, U. K.

Ethanol - BDH Limited, Poole, England, U. K.

3% Hydrogen peroxide - Sigma Chemical Company Limited, Poole, U. K.

Methanol - BDH Limited, Poole, England, U. K.

Goat serum - Scottish Antibody Production Unit, Carluke, U. K.

F(ab)2 fragments of goat anti-human IgG - Sigma Chemical Company Limited, Poole,

U. K.

123

Papain - Sigma Chemical Company Limited, Poole, U. K.

Biotin labelled fab fragments of goat anti-human lgG - Sigma Chemical Company


Biotin avidin HRP complex - Vector laboratories Inc., Burlingame, U. S. A

Developing solution - Vector laboratories Inc., Burlingame, U. S. A

Glycergel - Dako, Ely, Cambridgeshire, U. K.

Biotinamidocaproate N-Hydroxysuccimimidine ester (Biotin) - Sigma Chemical


2.3 General Buffers

Phosphate Buffered Saline (PBS): 8g NaCI, 0.2g KH2PO4,1.44g Na2HPO4.2H20,

0.2g KCI, dissolved in 800ml of distilled H2O. This was then made up to I litre and

stored at 4°C for a maximum of I week (pH 7.4).

Phosphate Buffered Saline containing 1mM disodium EDTA (PBSE) : 8g NaCl,

0.2g KH2PO43 1.44g Na2HPO4.2H20,0.2g KCI, were dissolved in 800m1 of distilled

H20.371mg EDTA was then added and the solution was made up to I litre and

stored at 4°C for a maximum of I week. The pH was adjusted when necessary with

5M NaOH to pH 7.4.

O. IM Sodium Carbonate Buffer containing 0.02% NaN3: 1.36g Na2CO3,7.35g

NaHCO3,200mg NaN3 were dissolved in 950m1 of distilled H2O. The pH was

adjusted to 9.2 using IM NaOH.

Tris Buffered Saline (TBS): 8g NaCI, 0.2g KCI, 3g Tris base were dissolved in

800m1 distilled H2O. 0.015g of Phenol red was added and the pH of the mixture was

124

adjusted to 7.4 with 1M HCI. Distilled H2O was then added to make the volume up to

1litre and the solution was stored at room temperature.

0.5% Sarkosyl: 0.5g w/v sodium lauryl sarcosinate (SDS) was dissolved in distilled

H20.

Coating Buffer (CB): 1.59g Na2CO3,2.93g NaHCO3 was dissolved in 800m1

distilled H20. The pH was adjusted to 9.6 at just under I litre, by adding IM HCl up

to 1 litre in a volumetric flask. It was stored in a sterilised bottle at 4°C for a

maximum of 1 week.

Incubation Buffer (IB): 8g NaCl, 0.2g KH2PO4,1.44g Na2HPO4.2H20,0.2g KCI,

0.5g Tween 20 dissolved in 800ml of distilled H20. Ig BSA was layered on the

surface and allowed to dissolve slowly. The volume was adjusted to I litre with

distilled H2O, then mixed and stored at 4°C for a maximum of 1 week (pH 7.4).

Wash Buffer (WB): This was prepared at 10 times the concentration of incubation

buffer (nil BSA) and stored at room temperature. It was diluted 1/10 immediately

before use.

N-tris[Hydroxymethyl]methyl-2-aminoethanesulfonic acid (TES) buffer: 229.2g

TES, 372.24g EDTA and 0.2g (w/v) SDS were dissolved in 100ml distilled H20. The

buffer was stored at 4°C for a maximum of 1 week.

Complete Medium: 50m1 foetal calf serum (FCS), 5ml penicillin (100001U), 5m1

streptomycin (10000µg/ml), 5ml L-Glutamine (200mM), 5m1 sodium pyruvate

(1mM), RPMI 1640 to 500m1.

CRL 1668 culture medium: 100m1 FCS, 10m1 L-Glutamine (400mM), G-418 at a

final concentration of 200µg/m1, Iscoves medium to 500m1.

125

Phosphate Buffered Saline containing 1mM disodium EDTA (Versene): 8g NaCI,

0.2g KH2PO4,1.44g Na2HPO4.2H20,0.2g KCI, dissolved in 800ml of distilled H20-

3.7 1g EDTA was then added and the solution was made up to I litre and stored at 4°C

for a maximum of 1 week. The pH was adjusted when necessary with 5M NaOH to

pH 7.4.

White cell counting fluid: 0.05% crytstal violet stain dissolved in 1% acetic acid.

Nigrosin in 7% acetic acid: 7m1 glacial acetic acid, 0.05g nigrosin, 0. lg sodium

azide dissolved in saline made up to a volume of 100ml. Filtration through filter

paper.

0.143M 2-aminoethylisothiouronium bromide (AET) solution: 0.402g of AET

dissolved in 10ml of distilled water, to pH 9.0 with 5M NaOH.

HAT (5x10-3M hypoxanthine, 2x10-5M aminopterin, 8x104M thymidine)

selection medium (HAT): I vial HAT concentrate dissolved in 10ml of RPMI 1640

medium, 100ml Foetal Calf Serum (FCS), 5ml glutamine (200mM), 5ml penicillin

(10000IU), 5m1 streptomycin (10000µg/ml), 1.25m1 transferrin (2mg/ml) 1.25m1

insulin (2mg/ml), Ouabain to final concentration lmM, then addition of RPMI 1640

to 500m1.

HT (5x10-3M hypoxanthine, 8x10-4 M thymidine) medium (HT): 1 vial HT

concentrate , dissolved in 10ml of RPMI 1640 medium, 100ml Foetal Calf Serum

(FCS), 5ml glutamine (200mM), 5m1 penicillin (10000IU), 5ml streptomycin

(10000µg/ml), 1.25m1 transferrin (2mg/ml), 1.25m1 insulin (2mg/ml), then addition of

RPMI 1640 to 500m1.

Phosphate buffered saline containing 0.1% tween-20 (PBS-T): 8g NaCl, 0.2g

KH2PO49 1.44g Na2HPO4.2H20,0.2g KCI, 0.5g Tween 20 dissolved in 800m1 of

distilled H20, and made up to I litre and stored at 4°C for a maximum of 1 week (pH

7.4)

126

Phosphate buffered saline containing tween and foetal calf serum (PBS-T-FCS):

8g NaCl, 0.2g KH2PO49 1.44g Na2HPO4.2H20,0.2g KCI, 0.5g Tween 20 dissolved in

800m1 of distilled H20, and made up to 1 litre. 5% FCS was then added. The buffer

was used on the day that it was prepared only.

O. 1M sodium acetate buffer (pH 4.8-5.0): 40.81g CH3COONa. 3H20 was dissolved

in 800ml distilled H2O. The pH was then adjusted with glacial acetic and the volume

made up to 1 litre with distilled H2O.

Chromogen substrate: I tablet of 3-amino-9-ethylcarbazole was dissolved in 2m1 N,

N-Dimethylformamide. 60m1 of 0.1M sodium acetate buffer, (pH 4.8-5.0) was then

added and the solution was filtered through a 0.20µm filter.

10% Buffered formaldehyde: 1 part buffered formaldehyde concentrate: 4 parts

distilled H20.

Mayers Haematoxylin: 2.5g of haematoxylin stain was dissolved in 2500ml of

distilled H2O using gentle heat. 125g of aluminium potassium sulphate was then

added and dissolved. 0.5g sodium iodate, 2.5g citric acid and 125g chloral hydrate

was then added, the solution was heated to boiling point for 5 minutes and then cooled

and filtered.

1% Eosin: 2.5g CaC12 was dissolved in 500ml of distilled H2O. 5g of eosin was then

added and dissolved.

Blocking serum: 120µl goat serum was added to 5ml of PBS.

Biotin avidin HRP complex mixture: 2 drops of solution A added to 5ml PBS, mix,

then 2 drops of solution B were added, mix. It is imperative that this solution is

prepared 30 minutes before use.

127

Developing solution: 2 drops of buffer added to 5m1 of distilled water, mix, 4 drops

of DAB, mix, 2 drops of H202, mix.

100mM phosphate buffer: 5.3g NaCl, 0.13g KH2PO45 0.96g Na2HPO4.2H20,0.13g

KCI, dissolved in 800ml of distilled H20. This was then made up to I litre and stored

at 4°C for a maximum of 1 week (pH 7.4).

128

Chapter 3. A Comparison of the Changes in the Humoral

Immune Response to Antigens of Periodontal Pathogens

Following Periodontal Therapy

3.1 Introduction

The existence of a humoral immune response in periodontal disease is a generally

accepted fact. Numerous studies have been carried out over the last few decades,

suggesting that elevations in the humoral immune response are associated with

increasing severity of disease (Genco et al., 1980b, Mouton et al., 1981, Tolo and

Brandtzaeg, 1982, Listgarten et al., 1981, Ebersole et al., 1982b, Naito et al., 1984,

Tew et al., 1985a). These studies suggest a relationship exists between specific

periodontopathic microbiota and antibody levels to the bacteria in both serum and

GCF.

The existence of a systemic humoral immune response to a variety of oral bacteria is

well documented in subjects with different forms of periodontal disease (Taubman et

al., 1982). This is probably simply because microbial challenge generally induces

antibody production and high titres indicate current or previous infection. One of the

major difficulties in evaluating the antibody response in periodontal disease has been

the antigenic complexity of periodontopathic bacteria. If investigations could lead to

the identification of major immunogenic proteins and antigens, and could characterise

the immune responses elicited more clearly, there may be a better understanding of the

host parasite interactions in periodontal disease.

The immune response has been studied in different forms of periodontal disease,

various age groups, and against a range of pathogens and antigens, and attempts have

been made to draw conclusions or detect differences. An important consideration is

the interaction between periodontal pathogens and the immune response of the host,

which is a dynamic one. Hence, temporal relationships and alterations may exist in

systemic antibody levels and relate to infection, treatment and active episodes of

disease. Indeed, epidemiological investigations have suggested that only a subset of

129

the population is susceptible to severe periodontitis, and while in some of these

individuals the disease appears to proceed as a chronic progressive destruction, in

others the disease presents itself in phases which can be periods of active or inactive

disease, or even remission (Socransky et al., 1994).

Although the nature of the humoral immune response is not fully understood, the host

response may provide a sensitive indicator of the flora found in the oral cavity,

particularly in the gingival sulcus (Tew et al., 1985b, Tew et al., 1985c, Vandesteen et

al., 1984). As well as providing information about the residing bacteria, the humoral

immune response has been suggested as a means of differentiating between distinct

periodontal disease states. High correlations between disease activity and antibody

titres have been reported for a number of micro-organisms. A high degree of

association has been reported for A. actinomycetemcomitans and localised aggressive

periodontitis (Altman et al., 1982, Astemborski et al., 1989, Ebersole et al., 1982b)

and P. gingivalis and chronic periodontitis (Mouton et al., 1981). Thus, diagnostic

differentiation of the periodontal disease forms has been suggested for both

periodontal microbiology and humoral immunology.

Due to the nature of periodontal disease, it is virtually impossible to obtain a serum

sample to investigate antibody targets and titres pre-disease. As a result, it is very

difficult to draw conclusions and identify associations between disease severity,

disease forms, antibody titres and micro-organisms. Even if a high correlation

between disease severity and antibody titre is identified, numerous subjects possess

antibody titres that fall above or below the ranges that identify health and disease, and

quite often this is irrespective of their periodontal health status. Antibody titres of

individuals not affected by periodontal disease ("normal") vary enormously. This

difficulty is compounded by the fact that most of the putative periodontal pathogens

can be found in healthy subjects (Loesche and Syed, 1978, Loesche et al., 1985) and

antibody titres against pathogens can be identified even in the absence of periodontal

disease (health) (Doty et al., 1982). Furthermore, it is often difficult to identify what

point in the disease process individuals are at, especially if they do not present with

disease until late on in life.

130

Due to all of these sources of variation between subjects, a number of studies have

investigated relative changes in antibody titres against various micro-organisms, by

assessing the success of a particular stage of periodontal therapy. These studies,

discussed below, attempt to monitor the microbial flora of periodontal patients by

assessing the host response. However, there are conflicting reports about the change

in antibody level after treatment.

Aukhil et al. (1988) observed serum antibody titres of patients with chronic

periodontal disease before and after periodontal treatment, against a number of oral

micro-organism sonicate preparations. They reported a general reduction in the mean

antibacterial titres following treatment. However, the intervals during which the

decreases in antibody titres occurred were not the same for each micro-organism

studied. Naito et al. (1987), reported a decrease in the serum antibody titre of

periodontally diseased patients to P. gingivalis post-treatment. A decrease in titre was

also reported by Mouton et al. (1987). This study reported the existence of two

subgroups within the periodontitis subjects studies. The distinction between the

subgroups was a difference in antibody reactivity against P. gingivalis. One subgroup

were IgA negative and had IgM and IgG titres not significantly higher than healthy

subjects. The second subgroup were IgA positive and showed IgM and IgG levels

much higher than those of the first subgroup. The latter group showed a 55% decrease

in antibody titres 5-7 months after treatment. Sandholm and Tolo (1987), also

observed serum antibody levels to 4 periodontal pathogens pre- and post-treatment in

aggressive periodontitis. Although they observed decreased antibody titres post-

treatment in 25% of the subjects tested, in the remaining 75%, they reported unaltered

antibody levels. More recently Horibe et al. (1995) reported a study on the effect of

periodontal treatments on IgG titres against 7 different periodontopathic bacteria. Of

the 7 bacteria tested, this group showed a significant decrease in antibody titres post-

treatment against P. gingivalis and P. intermedia. However, no difference in titres

between pre- and post-treatment were reported against Prevotella loescheii, F.

nucleatum, A. actinomycetemcomitans, Eikenella corrodens and Caphocytophga

ochracea.

131

The majority of the above studies report a decrease in antibody titre following

treatment, however, some results indicated no difference in titre pre- and post-

treatment against some of the bacteria tested. The studies reported are often carried

out for different lengths of time and are difficult to compare. Ebersole and Holt

(1991) noted that treatment can decrease serum IgG antibody levels to certain micro-

organisms, in both aggressive and chronic periodontitis patients, however, they

indicated that these changes were noted to take up to 8-10 months post-treatment and

were quite specific for selected micro-organisms in individual patients.

In contrast to these studies, reports have been made of investigations that have

indicated an increase in serum antibody titres to various micro-organisms following

treatment. Ebersole and Holt (1991b) reported that 60% of periodontitis patients

demonstrated increased titres against various periodontal pathogens post-treatment.

Sjöström et al. (1994) examined IgG titres to A. actinomycetemcomitans in 30 patients

with aggressive periodontitis. They reported a significant increase in titres at 6-12

months after beginning treatment. Mooney et al. (1995), also reported an increase in

antibody titre post-treatment. However, this study is slightly different and cannot be

directly compared with those above that suggest a decrease in titre following

treatment, as it observed the dynamics of the immune response to P. gingivalis and A.

actinomycetemcomitans in patients that were divided into subgroups. The group

divisions were based on the subject's baseline IgG antibody levels to P. gingivalis or

A. actinomycetemcomitans. The patients were labelled seropositive or seronegative.

Being labelled seropositive for a micro-organism meant the subject had an IgG titre to

an organism >2 times the control median. The study carried out by Mooney et al.

(1995), along with a similar study by Chen et al. (1991), added an extra factor that

included antibody avidity as well as antibody titre, and thus do not show antibody

titres post-treatment alone. Chen et al. (1991) looked at antibody titres and avidities

pre- and post treatment against P. gingivalis whole cell sonicates, LPS and protein

fractions. Again, in this study there were two groups, sero-positive and sero-negative

patients. The results indicated that following treatment, there was a decrease in

antibody titre against the antigens in the seropositive group and an increase in titre in

the sero-negative group. There was an increase in antibody avidity to the antigens

132

tested in both groups. Mooney et al. (1995), reported a significant increase in avidity

of IgG to P. gingivalis in the seropositive group, but no increase in titre (i. e. no

change). An increase in titre was reported for the sero-negative group. They also

reported a significant increase in titre against A. actinomycetemcomitans, but again in

the sero-negative group only. The studies reported by Mooney et al. (1995) and Chen

et al. (1991) therefore, suggest that the characteristics of the humoral immune

response to various pathogens may be quite different and depend on the

immunological status of the patient prior to treatment, a factor which may not have

been considered in many of the previously discussed studies. An increase in antibody

titre in seronegative patients could be due to the ìnoculation' theory suggested by

Ebersole et al. (1985d). Although the seronegative patients had a low antibody titre to

the antigens prior to treatment, these antibodies may have been a low level presence of

cross-reacting antibodies, and as a result of treatment, an antibody response may have

been mounted following induction of the primary immune response to these Gram

negative bacterial antigens. Chen et al. (1991) suggested that the increase in antibody

avidity in both groups of patients could be due to the treatment having an effect on

antibody responsiveness. They suggest that treatment leads to a reduction in antigenic

load, which in turn leads to a selection of clones of B lymphocytes that produce

antibodies of higher avidity.

The aim of the first part of the overall project was to investigate the humoral immune

response in a group of chronic periodontitis patients against a panel of whole micro-

organisms. In addition, responses to OMP and a sample of what are considered the

more potentially important antigens from these micro-organisms. The micro-

organisms used in this study were: P. gingivalis; A. actinomycetemcomitans; P.

intermedia; B. forsythus and T. denticola. The OMP of P. gingivalis, A.

actinomycetemcomitans, P. intermedia and B. forsythus were also used as well as the

antigens: HSP60 (of P. gingivalis); LPS (of P. gingivalis); leukotoxin (of A.

actinomycetemcomitans) and the ß segment of RgpA protease (of P. gingivalis). The

literature regarding these micro-organisms and antigens, and therefore, the reason why

they have been selected has been discussed already in the introduction to this thesis.

Although studies of this nature have previously been carried out, no study has yet

observed the titres against all of these targets in one study group so as to permit

133

comparisons. The study has been performed on a group of patients scheduled for

treatment for chronic periodontitis and antibody titres have been observed pre- and

post-treatment.

The second part of this study was investigating the cross-reactivity between a number

of these micro-organisms. Of the putative periodontal pathogens generally accepted

to play a role in the pathogenesis of periodontal disease, all are Gram negative bacteria

with the exception of the spirochaetes. Therefore, these different species will possess

a certain amount of homology, with structural features of surface components being

chemically very similar and possessing similar immunobiological properties. If these

surface components are immunogenic and play a role in the induction of the immune

response, perhaps there is some cross-reactivity occurring between antibodies induced

against one pathogen and a structurally very similar surface component of another

pathogen. Although numerous studies have been carried out to investigate serum

antibody titres against individual periodontopathogens, there does not appear to be a

large pool of literature regarding the question of cross-reactivity between the putative

periodontopathogens.

Evidence can be seen in the literature however, to suggest that this idea is worth

investigating. A study carried out by Kinane et al. (1993) observed specific

immunoglobulin titres to P. gingivalis and A. actinomycetemcomitans in the sera and

GCF of 20 patients with periodontitis. As part of the study they carried out

adsorptions of the sera against A. actinomycetemcomitans and Haemophilus

aphrophilus (H. aphrophilus). The adsorbed sera were then assessed for specific

antibody to A. actinomycetemcomitans. The analysis gave a mean reduction in the

anti-A. actinomycetemcomitans IgG of 68% and adsorption against H. aphrophilus

gave a similar reduction in the anti-A. actinomycetemcomitans IgG of 63%. Thus the

authors suggested that there are cross-reacting antibodies to these two species.

Moreover, the results of the study suggested a strong correlation between serum IgG

and IgM levels to these two bacteria, indicating cross-reactivity between antibodies

directed against these pathogens.

134

A study by Vasel et al. (1996) reported immunisation of monkeys with a vaccine

containing a monkey isolate of P. gingivalis that induced protection against alveolar

bone destruction. These workers investigated immune and pre-immune sera obtained from the monkeys, using ELISA and Western immunoblotting. The results of the

study indicated that following immunisation with killed P. gingivalis, the sera of the

monkeys contained increased antibody levels reactive with P. gingivalis and also B.

forsythus. Detailed analyses suggested that the binding of serum antibodies was a

specific, adaptive response to the LPS in the immunogen, since little reactivity was

detected in the pre-immune sera. Thus, the antibodies induced by LPS from P.

gingivalis appeared to be cross-reactive with the LPS of B. forsythus.

Further to the above study other investigations have suggested cross-reactivity

between antibodies against LPS in non-oral Gram negative bacteria. Seifert et al.,

(1996) reported the generation of a human monoclonal antibody that recognised a

conserved epitope shared by LPS molecules of different Gram negative bacteria,

indicating a high degree of cross-reactivity of antibody across different bacterial

species.

Hinode et al. (1998) reported a clear cross-reactivity between antibodies raised to

GroEL-like proteins of P. gingivalis, A. actinomycetemcomitans, and E. coli,

indicating that a significant portion of the antibodies induced against GroEL-like

proteins are directed at epitopes conserved across bacterial genera.

Thus, there is evidence indicating cross-reactivity between antigens of different Gram

negative bacteria. The aim of the second part of this serum antibody study was to

investigate cross-reactivity between 4 periodontal pathogens, P. gingivalis, A.

actinomycetemcomitans, P. intermedia and B. forsythus. Serum was used that had

been taken from 9 individuals suffering from chronic periodontitis and was adsorbed

against one bacteria before the antibody titres against all 4 pathogens was tested in an

ELISA.

Previous adsorption studies that have been undertaken mainly involving only two

micro-organisms (Vasel et al., 1996; Kinane et al., 1993; Seifert et al., 1996). It was

135

felt that this only revealed part of the cross-reactivity story and that additional cross-

reactivity testing would provide more information on P. gingivalis, A.

actinomycetemcomitans, B. forsythus and P. intermedia. This investigation is

therefore, of interest as the extent of cross-reactivity between more than two putative

periodontopathogens is being investigated in a single study.

3.2 Methods Part 1: Enzyme Linked Immunosorbent Assay

3.2.1 Subjects and blood samples

Prior to the commencement of these studies, ethical approval was obtained from the

Glasgow Dental Hospital Ethics Committee. Subjects participating in these studies

were informed of the protocol and consent was obtained. All patients taking part in

the investigation were free to withdraw from the study at any time. Twenty five

subjects, 11 males and 14 females, with a mean age of 47.3 + 7.6 years (mean + S. D

(range = 36-66 years), were enrolled in the study. Seven of the subjects were

smokers, 18 were non-smokers. The subjects were patients attending Glasgow

University Dental Hospital. The criteria for inclusion was that patients had advanced

periodontal disease with probing depths (P. D. ) of >5.0 mm in at least 2 non-adjacent

sites per quadrant. Site selection was carried out at the screening visit, where full

mouth periodontal pocket charting was performed using a PCP-12 periodontal probe.

Wherever possible, one site in each quadrant over 5 mm in depth, without furcation

involvement, was selected for inclusion in the study. These subjects had no history of

systemic conditions which could influence the course of periodontal disease, had not

received antibiotic treatment during the previous 3 months, and had received no prior

periodontal treatment.

At the baseline appointment, P. D. and bleeding on probing (BOP) at these sites were

assessed with the Florida probe (Gibbs et al., 1988) and blood samples taken. Twenty

one millilitres of venous blood were collected from the anterior cubital region using

butterfly needles and 3x7 ml non-heparinised vacutainer tubes. The blood was

allowed to clot overnight at 4°C. It was then centrifuged at 750 xg for 15 minutes and

the serum was separated, aliquoted and stored at -80°C until required for ELISA

136

analysis. Patients were treated by an experienced periodontist and given a course of

Hygiene Phase Therapy (HPT). This included oral hygiene instruction, quadrant

scaling and root planing under local anaesthesia.

The subjects were reassessed with conventional probing charts after HPT. A blood

sample was taken and P. D. and BOP measurements were recorded at the 4 pre-

selected sites with the Florida probe 138.9 +55.4 days (mean + S. D. ), range = 56 - 280

days, after initial sampling.

3.2.2 Antigen preparation

1) Whole fixed Bacteria

P. gingivalis NCTC 11834 and P. intermedia ATCC 25611 were grown under

anaerobic conditions (85% N29 10% H2,5% C02) on fastidious anaerobe agar (FAA).

B. forsythus ATCC 43037 was also grown under anaerobic conditions in tryptic soy

agar supplemented with 7% horse blood and 10µg/ml N-acetylmuramic acid. A.

actinomycetemcomitans ATCC 29523 (serotype A) was grown in 5% CO2 and 95%

air on columbia blood agar. All bacteria were grown at 37°C. A.

actinomycetemcomitans was harvested after 48 hours, P. gingivalis and P. intermedia

after 5 days and B. forsythus after 7 days. T. denticola ATCC 35405 were cultured in

medium OMIZ-pat + 1% human serum, and were a kind gift from Dr. C. Wyss, Institut

fur Orale Mikrobiologie, Zurich, Switzerland (Wyss et al., 1997).

All bacterial cells were harvested into phosphate buffered saline containing 1mM

disodium EDTA (PBSE), washed by centrifugation at 200 xg for 10 minutes, and

fixed for 1 hour at room temperature in 10% formal saline. The cells were then

washed twice in PBSE and once in 0.1 M sodium carbonate-bicarbonate buffer

containing 0.02% NaN3 at pH 9.6. Fixed cells were stored at 4°C until use.

137

2) Outer-membrane Protein Preparation

The preparation of the outer-membrane antigens was carried out using a method

modified from that of Schryvers et al. (1988). P. gingivalis, A.

actinomycetemcomitans, P. intermedia and B. forsythus were grown as described

above. The bacteria were harvested into PBSE and centrifuged at 10,000 xg for 20

minutes. The pellet was then resuspended in tris buffered saline (TBS) and washed a

further twice, as described above before being resuspended. The cells were sonicated

for 10 pulses at 200W for 10 seconds, with a 2-minute cooling period between each

pulse. Centrifugation was then carried out for 15 minutes at 8000 rpm at 4°C. The

supernatant was removed and transferred into sterile ultracentrifuge tubes.

Ultracentrifugation was then carried out for 1 hour at 40,000 rpm at 4°C and the pellet

resuspended in 0.5% sarkosyl and retained at 4°C overnight. Ultracentrifugation was

then carried out as above and the pellet resuspended in distilled water.

The protein concentrations were determined using the BCA protein assay kit.

3) Purified Antigen Preparation

Heat Shock Protein 60 of P. gingivalis

The heat shock protein 60 (HSP60) (of P. gingivalis) was purified according to Maeda

et al. (1994), and was a kind gift from Dr. H. Maeda, Department of Periodontology

and Endontology, Okayama University Dental School, Okayama, Japan.

Leukotoxin of A. actinomycetemcomitans

The leukotoxin (of A. actinomycetemcomitans) was purified in accordance with the

protocol reported by Tsai et al. (1984) with modifications and was a kind gift from Dr.

J. Korostoff, Leon Levy Research Center for Oral Biology, University of

Pennsylvania, Philadelphia, USA.

Lipopolysaccharide of P. gingivalis

Lipopolysaccharide of P. gingivalis W50 was purified according to the method

described by Darveau and Hancock (1983) and was a kind gift from Professor M.

138

Curtis, Department of Oral Microbiology, Medical research Council Pathogenesis

Group, St. Bartholomews and The Royal London School of Medicine and Dentistry,

London, UK.

The (3 segment of the RgpA protease of P. gingivalis

Domains of the RgpA protease of (P. gingivalis) were prepared as an N-terminal

(His)6 fusion protein as described by Slaney et al. (2000) and was a kind gift from

Professor M. Curtis.

3.2.3 Analysis of sera samples by Enzyme Linked Immunosorbent Assay (ELISA)

Specific antibody titres were measured using enzyme-linked immunosorbent assay

(ELISA) by the method of Ebersole et al. (1980).

The plates were pre-washed 3 times with coating buffer and coated overnight with the

antigen at 4°C using 100µl per well. Immulon lB plates were used because of their

low protein-binding characteristics and low non-specific background. The antigens

were coated onto the surface of the wells at a concentration which had been

determined as optimum to coat these plates. These concentrations were P. gingivalis

and P. intermedia whole cells; 0.05 at OD600, A. actinomycetemcomitans and B.

forsythus whole cells; 0.02 at OD600, T. denticola whole cells; 0.001 at OD600, P.

gingivalis and A. actinomycetemcomitans OMP; 10µg/ml, P. intermedia and B.

forsythus OMP; 0.5µg/ml, lipopolysaccharide (P. gingivalis); 4µg/ml, leukotoxin (A.

actinomycetemcomitans); 0.5µg/ml, HSP60 (P. gingivalis) and RgpA (3 segment (P.

gingivalis); 2µg/ml. Control wells for each aspect of the ELISA process were

arranged around the outside of the plates.

The ELISA plates were removed from the refrigerator and washed 4 times, then 4

more, 4 times again and lastly once more with wash buffer (4x4x4x 1). The non-

specific binding sites were blocked with 100µl per well of incubation buffer (IB)

containing 5% skimmed milk powder for 1 hour at 37°C. The plates were then

washed twice and once again (2x 1) before addition of the sera.

139

Human sera collected from the study patients were diluted 1/200 using 113.50µl of

sera were added to each well and incubated at 37°C for 90 minutes. Each serum was

tested in duplicate and the sera samples from the same patient at different time points

were tested on the same plate. Following incubation with the sera the plates were

washed 4x4x4xl. Afterward the plates were incubated at 37°C for 60 minutes with

Biotin-goat-anti-human IgG at 1/2000 dilution in IB. The plates were then washed

4x4x4x1 and then incubated for 60 minutes at 37°C with 1/2000 dilution of extravidin

peroxidase. After washing 4x4x4xl, the reactions were visualised using 100µ1 per

well of TMB Peroxidase substrate and stopped after 5-10 minutes, depending on the

speed of the reaction, using 50pl per well of 0.12M HC1.

The reactions were read using a Dynex Technologies MRX II plate reader at 630nm

with a reference at 490nm and the results printed out. The duplicate results were

averaged and the nonspecific background binding was subtracted. The final titre was

expressed as O. D. units as has been done in numerous other studies (Suzuki et al.,

1984; Murray et al., 1989; Sandholm and Tolo, 1987; Cridland et al., 1994). The raw

data was analysed in this study as the data being compared was obtained from the

same 96 well plate in every case. Therefore, as all titres being compared were

obtained at the same time point, it was not necessary to use a standard curve to correct

for different variables. This study is comparative, rather than quantitative, therefore,

the units are arbitrary. As such, converting the data into ELISA units would be an

unnecessary data processing step.

3.2.4 Statistical methods

The data obtained from this study revealed that the values were adequately normalised

by blotplots, although in a number of cases there were a few outliers. As the data was

normalised, analyses of mean responses were undertaken using the statistical package

Minitab, and parametric tests were used to assess relationships. Means, standard

deviations and ranges are presented.

Mean values for pocket depths of the subjects were used. Significance levels for all

tests were p< _ 0.05, unless otherwise stated.

140

3.3 Methods Part 2: Adsorptions

3.3.1 Subjects and blood samples

Prior to the commencement of these studies, ethical approval was obtained from the

Glasgow Dental Hospital Ethics Committee. All patients taking part in the

investigation were free to withdraw from the study at any time. Nine Subjects, 5

males and 4 females, were enrolled in this study. The subjects were patients attending

Glasgow University Dental Hospital and they had a mean age of 45.2 + 11.3 years

(mean + S. D. ) (range = 29-60 years). The criteria for inclusion was as previously

reported 3.2.1.

The blood was collected and stored as reported in 3.2.1.


P. gingivalis NCTC 11834, P. intermedia ATCC 25611, B. forsythus ATCC 43037

and A. actinomycetemcomitans ATCC 29523 (serotype A) were grown as described in

3.2.2

All bacterial cells were harvested into PBSE, washed by centrifugation at 200 xg for

10 minutes. The bacterial pellet was then resuspended in PBS containing TLCK at

50µg/ml and Leupeptin at 0.1mg/ml. The cells were sonicated on ice using a

Microson Ultrasonic Cell Disrupter for a1 minute period, with a 1-minute cooling

period following. This was repeated 10 times. The cell suspension was then

microcentrifuged for 10 minutes at 13,000 rpm, before the supernatant was discarded

and the cell pellet was resuspended in PBS again containing TLCK and leupeptin, at

the same concentrations as described above.

141

3.3.3 Adsorption of samples

The plates were pre-washed 3 times with TES buffer and coated overnight with the

antigen at 4°C using 100µl per well. Immulon IB plates were used because of their

low protein-binding characteristics and low non-specific background. Sonicates of P.

gingivalis, A. actinomycetemcomitans, P. intermedia and B. forsythus were coated

onto the surface of the wells at a concentration of 10µg/ml, which had previously been

reported to be optimal (Vasel et al., 1996). The sonicate preparations were diluted in

CB.

The following day the plates were washed with WB before the addition of human

serum diluted 1/100 in TES buffer. The plates were incubated for 1 hour with

agitation at 37°C, before the serum was transferred to coated wells that had not been

used and incubated for a further hour at 37°C. This was repeated 5 times. The serum

was then moved to a fresh batch of pre-coated wells and the plate was incubated at

4°C overnight.

The following day, the serum was again moved to a fresh batch of pre-coated wells

and incubated for 1 hour at 37°C with agitation, this was repeated 4 times to give a

total of 10 x1 hour incubations.

The serum was then removed from all of the wells and stored at 4°C until use.

The whole experiment was carried out in duplicate but instead of coating the plates

with antigen they were coated with CB alone. This was so all of the serum samples

could be sham absorbed as well as absorbed against the antigens.

3.3.4 Analysis of sera samples by Enzyme Linked Immunosorbent Assay (ELISA)

Serum antibody titres were measured using ELISA as described in 3.2.3. Each sample

was assayed in triplicate.

142

The antigens were coated onto the surface of the wells at a concentration of 10µg/ml.

Controls for each aspect of the ELISA process were arranged in the wells around the

outside of the plates.

Human sera, both absorbed and sham-absorbed were used at 1/100 dilution.

The triplicate results were averaged and the final titre expressed as O. D units.

3.4 Results

3.4.1 Results of part 1 of the study: Enzyme Linked Immunosorbent Assay

The mean age of the subjects was 47.3 + 7.6 years (mean + S. D. ). The length of

treatment that each individual experienced varied quite extensively. The mean length

of time between the taking of the first and second samples was 138.9 + 55.4 days

(mean + S. D. ), however, the range was 56 to 280 days, which makes it difficult to use

the length of treatment parameter in statistical analyses.

143

Table 3.1 Polyvalent immunoglobulin titres for before and after treatment and

the differences between these (O. D. units)

Mean S. D. Mean S. D. Mean S. D. 95% CI of p- BT AT Difference Difference value*

(BT-AT) (BT-AT) PD 5.99 1.00 4.37 1.20 1.62 0.98 (1.22, <0.001

2.02) Pg 0.89 0.27 0.90 0.27 0.00 0.09 (-0.04, 0.87

0.04) Aa 0.32 0.22 0.34 0.23 0.02 0.14 (-0.08, 0.47

0.04) Pi 0.82 0.25 0.79 0.24 0.03 0.10 (-0.01, 0.14

0.07) Bf 0.23 0.10 0.22 0.10 0.01 0.11 (-0.04, 0.63

0.06) Td 0.28 0.12 0.25 0.13 0.03 0.10 (-0.01, 0.16

0.07) Pg OMP 0.52 0.19 0.51 0.17 0.02 0.07 (-0.01, 0.18

0.05) Aa OMP 0.72 0.14 0.73 0.15 0.01 0.07 (-0.03, 0.55

0.02) Pi OMP 0.73 0.24 0.73 0.22 0.00 0.10 (-0.04, 0.91

0.04) Bf OMP 0.46 0.15 0.46 0.12 0.00 0.10 (-0.04, 0.82

0.05) Aa LTX 0.75 0.39 0.75 0.38 0.00 0.11 (-0.04, 1.0

0.04) Pg 0.25 0.14 0.25 0.12 0.01 0.06 (-0.02, 0.63 LPS 0.03) Pg 0.24 0.12 0.23 0.09 0.01 0.06 (-0.01, 0.39 HSP60 0.04) Pg RgpA 0.18 0.10 0.16 0.12 0.02 0.10 (-0.02, 0.34

0.06)

*p-value for hypothesis test of no difference between before treatment and after

treatment (paired t-test).

BT - before periodontal treatment, AT - after periodontal treatment, PD - pocket

depth, Pg - P. gingivalis, Aa - A. actinomycetemcomitans, Pi - P. intermedia, Bf - B.

forsythus, Td - T. denticola, Ec - E. coli, OMP - outer-membrane proteins, Aa LTX -

Leukotoxin (of A. actinomycetenicomitans), Pg LPS - Lipopolysaccharide (of P.

144

gingivalis), Pg HSP60 - Heat shock protein 60 (of P. gingivalis), Pg RgpA -P

segment of RgpA protease (of P. gingivalis).

145

Figure 3.1 Plot of polyvalent immunoglobulin titres pre- and post-treatment

against P. gingivalis

1.4

1.3

1.2

a, 1.1

1.0

-ö 0.9 0

0.8

0.7

0.6

0.5

0.4

C)

After > Before (Increase) ""

"

"""

""

" ""

"

" " ""

""

" " Before > After (Decrease)

0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3

Before Treatment Antibody Titre (ELISA units)


against P. gingivalis OMP

, -. 1.0

After> Before (Increase) 0.9

0.8 "

0 7 "

" .

0.6 "

0.5 .U"

0.4 " " """"

"

0.3 "

2 " Before > After (Decrease)

w. 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.

Before Treatment Antibody Titre (O. D. units)

146


against the ß segment of RgpA (of P. gingivalis)

-.

4 93 0.7

0.6 W

0.5

F" 0.4

0.3

0.2

0.1

F; 0.0

d

After > Before (Increase) "

"

" "

1=M """ ". " ""

Before > After (Decrease)

0.0 0.1 0.2 0.3 0.4 0.5


147

Table 3.1 shows the data collected for the pocket depths recorded pre- and post-

treatment and the changes in these following treatment. Again the ranges of pocket

depths were quite large both pre-and post-treatment, however, statistical analysis

shows that there is a highly significant decrease in pocket depth following treatment.

Table 3.1 also shows the data collected following analysis of the serum by ELISA

against the panel of micro-organisms antigen preparations before and after treatment.

The polyvalent immunoglobulin titres (IgA, IgG and IgM) are expressed in O. D. units.

Statistical analyses of these results indicate that there was no significant change in

antibody titre following treatment for antibody levels against any of the micro-

organisms or antigens tested. When the data is analysed in graphical form for each of

the micro-organisms and antigens for a number of the targets outliers can be observed.

Figure 3.1 shows the plot of polyvalent immunoglobulin titres pre- and post-treatment

against P. gingivalis. For each of the different analyses carried out, the graph for

antibody titres for P. gingivalis will be used to show a representative sample of

results. Other graphs are only shown to indicate a point of interest or a significant

result.

Figure 3.2 shows the plot of polyvalent immunoglobulin titres pre- and post treatment

for the 25 patients against the OMP of P. gingivalis and it can be seen that there are

two outliers, patients 12 and 18. If these two patients are excluded from the statistical

analysis, there is a significant decrease in antibody titre following treatment against

this antigen preparation (mean difference = 0.021, S. D. = 0.046,95% CI = 0.00,0.04,

p-value = 0.04). However, there seems to be no valid reason either clinically or

experimentally, to exclude these 2 patients from the analysis and hence, the overall

results indicate that there is no significant change in antibody titre following

treatment.

Figure 3.3 which shows the plot polyvalent immunoglobulin titre pre- and post-

treatment against the 1 segment of RgpA (of P. gingivalis), also indicates an outlier.

When this subject is removed from the statistical analysis, a significant decrease in

titre can be seen following treatment (mean difference = 0.03, S. D. = 0.07,95% CI =

148

0.00,0.06, p-value = 0.03), however, again there is no valid reason to exclude this

subject from the analysis, and hence the overall data suggests no change in antibody

titre against this antigen following treatment.

149

Figure 3.4 Boxplot of the difference between pre- and post-treatment antibody

titres against LPS

O

b 0

0.1 * * *

.ý

6 E_

Q

0.0

*

-0.1

-0.2 *

150

Figure 3.4 shows a box plot for the difference in antibody titres against LPS before

and after treatment. It can be seen that 6 outliers appear to be slightly skewing the

data. When all 6 of the subjects are removed from the analysis, there is still no

significant difference between pre- and post treatment antibody titres. On removal of

the 6 outliers the p-value for the difference in titres becomes larger. For a few of the

micro-organisms and antigens, the boxplots showed a number of outliers, however,

removal of them did not lead to a significant result, except for the incidents of P.

gingivalis OMP and RgpA as discussed previously.

Patient I appeared to be an outlier on analysis of antibody titre change against a

number of different micro-organisms and antigens. This patient always showed a

decrease in antibody titre following treatment. The reason for this is not clear. It is

noted, however, that patient (number) 1 had a pre-treatment pocket depth of 5.65mm

and a final pocket depth of 5.45mm indicating that any decrease in pocket depth and

improvement was insignificant. However, no conclusions can be made from this

individual alone, and there was no significant correlation between change in pocket

depth and change in antibody titre following treatment when all subjects were

observed together (see table 3.3 below).

151

Table 3.2 Polyvalent immunoglobulin, IgG and IgA titres before and after

treatment and comparisons of these for P. gingivalis, A. actinomycetemcomitans,

P. intermedia, B. forsythus and T. denticola

Mean BT

S. D. Mean AT

S. D. Mean Difference (BT-AT)

S. D. 95% CI of Difference (BT-AT)

p- value

Pg Poly Ig 0.89 0.27 0.90 0.27 -0.01 0.09 (-0.04,0.04) 0.87 IG 1.03 0.35 1.05 0.36 0.02 0.08 (-0.05,0.02) 0.35 IgA 0.34 0.29 0.35 0.28 -0.01 0.07 (-0.03,0.02) 0.74 Aa Poly Ig 0.32 0.22 0.34 0.23 0.02 0.14 (-0.08,0.04) 0.47 IgG 0.32 0.28 0.35 0.31 0.03 0.10 (-0.01,0.07) 0.11 IgA 0.06 0.06 0.06 0.06 0.00 0.03 (-0.01,0.01) 0.89 Pi Pol Ig 0.82 0.25 0.79 0.24 0.03 0.10 (-0.07,0.01) 0.14 IgG 0.86 0.30 0.85 0.28 0.01 0.10 (-0.05,0.03) 0.65 IgA 0.16 0.10 0.16 0.10 0.00 0.07 (-0.03,0.03) 0.85 Bf Poly Ig 0.23 0.10 0.22 0.10 0.01 0.11 (-0.06,0.04) 0.63 IG 0.14 0.08 0.14 0.08 0.00 0.07 (-0.03,0.02) 0.76 IA 0.06 0.05 0.06 0.04 0.01 0.04 (-0.02,0.01) 0.50 Td Poly Ig 0.28 0.12 0.25 0.13 0.03 0.10 (-0.01,0.07) 0.16 1G 0.20 0.09 0.17 0.08 0.03 0.09 (-0.07,0.01) 0.15 IgA 0.07 0.05 0.06 0.03 0.01 0.04 (-0.03,0.01) 0.31

* p-value for hypothesis test of no difference between before treatment and after

treatment (paired t-test).

152

Figure 3.5 Plot of IgG pre- and post-treatment against P. giizgivalis

1.5 d

ky

1.0 0

i-y

0.5

N

cz d

After > Before (Increase) " bee 0

"

" "

"

" "

" "

"""

0 Before > After (Decrease)

0.5 1.0 1.5


Figure 3.6 Plot of IgA titres pre- and post-treatment against P. gingivalis

1.0

0.5

.r cz aý Fý 0.0 a)

After> Before (Increase)

a

S ""

"" "" ""

"

6

Before > After (Decrease)

"

"

0.0 0.5 1.0


153

Table 3.2 shows the results for the titres of polyvalent immunoglobulin, IgG and IgA

against P. gingivalis, A. actinomycetemcornitans, P. interinedia, B. forsythus and T.

denticola pre- and post-treatment. The mean highest antibody titres were either

polyvalent or of the isotype IgG, and the lowest IgA in most cases. No significant

differences between any of the isotypes (polyvalent, IgG and IgA) were noted

following treatment. Figure 3.5 shows the plot of the IgG titres pre-and post-

treatment against P. gingivalis. Figure 3.6 shows the plot of the IgA titres pre- and

post-treatment against P. gingivalis.

154

Table 3.3 The correlation between change in pocket depth and change in

antibody titre following treatment

Correlation between change in pocket depth (mm) and change in antibody titre

(O. D. units) against:

Correlation coefficient

95% Cl for correlation coefficient

p-value*

Pg -0.073 (-0.46,0.33) 0.728 Aa 0.300 (-0.11,0.62) 0.146 Pi -0.099 (-0.48,0.31) 0.639 Bf 0.566 0.22,0.79) 0.003 Td -0.040 (-0.43,0.36) 0.849 Pg OMP 0.063 (-0.34,0.45) 0.766 Aa OMP -0.059 (-0.44,0.34) 0.780 Pi OMP -0.122 (-0.49, -0.29) 0.562 Bf OMP -0.197 (-0.55,0.22) 0.346 Aa LTX -0.204 (-0.55,0.21) 0.328 Pg LPS -0.105 (-0.48,0.30) 0.616 P HSP60 0.133 (-0.28,0.50) 0.527 Pg RgpA 0.245 (-0.17,0.58) 0.239

*p-value for hypothesis test of no correlation/correlation coefficient equal to zero

(Pearson product moment coefficient of correlation)

155

Figure 3.7 Plot of change in pocket depth against change in antibody titre for B.

forsythus

0.2

0.1

0.0

H -0.1

0

-0.3

-0.4

-

.. . .

. . .

.

.

0123

Change in Pocket Depth (mm)

Figure 3.8 Plot of change in pocket depth against change in antibody titre for P.

gingivalis

0.2

0.1

1

0.0 0

-0.1

N on

-0.2

f _

f "

" "

0123

Change in Pocket Depth (mm)

156

Table 3.3 shows the correlation between change in pocket depth and antibody titre

against the micro-organisms and antigens tested following treatment. There is a

significant correlation (non zero correlation) between change in pocket depth and

antibody titre against B. forsythus, post-treatment. The correlation coefficient for this

relationship is 0.566, implying that as the change in pocket depth increases, the

change in antibody titre also increases. The graph suggests that the larger the decrease

in pocket depth following treatment, the greater is the positive change in titre and

hence the greater the increase in titre. Figure 3.7 shows that there is a correlation

between change in antibody titre to B. forsythus and change in pocket depth. The

graph is further evidence of a correlation. A positive correlation, although not

significant can also be seen for change in pocket depth and change in titre against A.

actinomycetemcomitans. Therefore, it is difficult to explain why there is a positively

skewed correlation for A. actinomycetemcomitans and B. forsythus whole cells, but

not for A. actinomycetemcomitans and B. forsythus OMP or A.

actinomycetemcomitans leukotoxin.

One might remark that since 13 comparisons were made against pocket depth, i. e. one

comparison for each antigen, some investigators might suggest the Bonferroni's

approach (Glantz, 1981) to "correct" the individual p-values to ones more appropriate

for the situation in which multiple, related tests are performed. Therefore, the only

results that are significant in table 3.3 are those for which the uncorrected p-values are

< 0.015. As the only significant result of change in pocket depth against change in

antibody titre to B. forsythus has a p-value of 0.003, the Bonferroni's approach has no

effect in this case. Figure 3.8 shows the plot of change in pocket depth against change

in antibody titre against P. gingivalis.

157

Table 3.4 To show the correlation between pre-treatment pocket depth and pre-

treatment antibody titre

Correlation between pre-treatment

pocket depth (mm) and pre-treatment

antibody titre (O. D. units) against:

Pearson product

Moment

Correlation

coefficient

95% CI for

correlation

coefficient

p-

value*

Pg 0.465 ( 0.09,0.73) 0.019

Aa 0.573 ( 0.23,0.79) 0.003

Pi 0.149 (-0.26,0.51) 0.477

Bf -0.219 (-0.57,0.19) 0.293

Td -0.074 (-0.46,0.33) 0.725

Pg OMP 0.333 (-0.07,0.64) 0.104

Aa OMP 0.388 (-0.01,0.68) 0.055

Pi OMP 0.054 (-0.35,0.44) 0.799

Bf OMP -0.026 (-0.42,0.37) 0.900

Aa LTX 0.339 (-0.06,0.65) 0.097

Pg LPS 0.332 (-0.07,0.64) 0.104

Pg HSP60 0.057 (-0.35,0.44) 0.786

Pg RgpA -0.291 (-0.62,0.12) 0.158


(Pearson product moment coefficient of correlation)

158

Figure 3.9 Plot of pre-treatment pocket depth against pre-treatment antibody

titre for A. actinomycetemcomitans

Hfl

H

0

H

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

" "

" " " " "

" " " " f " "

"" " "

56789 Pre-Treatment Pocket Depth (mm)

159

Table 3.4 shows the correlation between pre-treatment pocket depth and pre-treatment

antibody titre. Statistical analysis has indicated there is significant correlation (non

zero correlation) between pre-treatment pocket depth and pre-treatment antibody titres

against P. gingivalis, A. actinomycetemcomitans and A. actinomycetemcomitans OMP

preparation. The correlation coefficient for P. gingivalis is 0.465 and 0.573 for A.

actinomycetemcomitans. Although, the p-values are significant (0.019 and 0.003

respectively), the correlation's are not strong. The correlation coefficient for pre-

treatment pocket depth and pre-treatment antibody titre against A.

actinomycetemcomitans OMP preparation is not significant. The correlation

coefficient is 0.388 and the p-value is 0.055, however, the confidence interval is -0.01, 0.68. This indicates that there is a correlation, however, this is not significant. All 3

of the correlation's are positive, indicating that increased pre-treatment pocket depths

can be correlated with increased pre-treatment antibody titres against P. gingivalis, A.

actinomycetemcomitans and A. actinomycetemcomitans OMP preparation. Figure 3.9

shows the plot of pre-treatment pocket depth against pre-treatment antibody titre to A.

actinomycetemcomitans. It can be seen from the graph that there are 2 outliers. When

examined these patients are not young, i. e. there is no chance that they may be

suffering from aggressive periodontitis instead of chronic periodontitis, and therefore,

the reason for them being outliers is not clear.

Again some investigators may suggest that the Bonferroni's approach to "correct" the

individual p-values would be appropriate in this situation. If it is used, then the p-

value is < 0.015 which means that only the correlation between pre-treatment pocket

depth and antibody titre against A. actinomycelemcomitans is significant.

160

Table 3.5 To show the correlation between length of treatment and change in

antibody titre following treatment

Correlation between length of treatment

and Change in antibody titre following

treatment against:

Correlation

coefficient

95% Cl for

correlation

coefficient

p-value*

Pg -0.102 (-0.48,0.31) 0.629

Aa 0.008 (-0.39,0.40) 0.969

Pi 0.241 (-0.17,0.58) 0.245

Bf -0.201 (-0.55,0.21) 0.335

Td -0.032 (-0.42,0.37) 0.879

Pg OMP 0.280 (-0.13,0.61) 0.175

Aa OMP 0.218 (-0.19,0.56) 0.296

Pi OMP -0.005 (-0.40,0.39) 0.980

Bf OMP 0.174 (-0.24,0.53) 0.404

Aa LTX 0.092 (-0.31,0.47) 0.663

Pg LPS 0.003 (-0.39,0.40) 0.988

Pg HSP60 -0.522 (-0.76, -0.16) 0.007

Pg RgpA -0.101 (-0.48,0.31) 0.632


(Pearson product moment coefficient of correlation).

161

Figure 3.10 Plot of length of treatment against change in antibody titre for heat

shock protein 60 (of P. gingivalis)

0.1

H

0.0

b4

-0.1

.

". "" ". "

" ""

"""

i0 150 250

Length of Treatment (days)

Figure 3.11 Plot of length of treatment against change in antibody titre for P.

gingivalis

0.2

0.1

0.0

0.1

on

-0.2 U

" "6"

-"

U " ""

" "

""" "'

""

"

.I

50 150 250

Length of Treatment (days)

162

The correlation between change in antibody titre following treatment against various

micro-organisms and antigens and the length of treatment undergone by the subjects is

shown in table 3.5. There is a significant correlation recorded (non zero correlation)

for the change in antibody titre following treatment against HSP60 (of P. gingivalis)

and the length of treatment experienced by the subjects. The correlation coefficient is

-0.522 implying that as the length of treatment increases, the change in antibody titre

decreases. This relationship can be seen more clearly in figure 3.10. This figure

shows the plot of length of treatment against the change in antibody titre to HSP60 (of

P. gingivalis). It can be seen from the graph that there is a trend towards a negative

correlation and that if the treatment is shorter than approximately 125 days, the change

in antibody titre is positive (above zero) i. e. an increase in antibody titre following

treatment. A period of treatment of above 125 days or more seems to lead to a change

in antibody titre that is negative (below zero) i. e. a decrease in antibody titre following

treatment. This correlation is still significant following application of the

Bonferroni's approach. Figure 3.11 shows the plot of length of treatment against

change in antibody titre to P. gingivalis.

163

3.4.1.1 Relationship between treatment and antibody titres

An analysis of the 25 patients before and after treatment can be seen in tables 3.1 - 3.6. The analyses looked at the effect of treatment on polyvalent immunoglobulin

titres against a large panel of bacteria and antigens, and 3 different isotypes of

immunoglobulin; polyvalent immunoglobulin, IgG and IgA against 5 whole putative

periodontopathogen preparations. The analyses suggests that the treatment did not

have an effect on antibody titres against the micro-organisms and antigens tested in

this study and that there was no significant difference between antibody titres before

and after treatment.

The data in table 3.1 shows the highest titres of polyvalent immunoglobulin were

specific for P. gingivalis and P. intermedia whole fixed bacteria, and the antibody

titres against purified antigens such as LPS, HSP60 and the ß segment of RgpA were

significantly lower.

Levels of 3 different isotypes of immunoglobulin were investigated before and after

treatment for 5 whole fixed putative periodontopathogens. Analysis showed no

significant difference between pre- and post-treatment titres for any of the isotypes

against any of the micro-organisms. The IgA titres before and after treatment were

extremely low when compared with the titres of polyvalent immunoglobulin and IgG.

The levels of polyvalent immunoglobulin and IgG were very similar, and in some

cases the IgG levels were slightly, but not significantly, higher than the levels of

polyvalent immunoglobulin.

Although the results are not presented, analysis did show that the change in antibody

titre against most of the micro-organisms and antigens tested correlated significantly

with their corresponding pre-treatment antibody titre. The correlations were all

positive, indicating that an increased pre-treatment titre against a particular antigen

can be correlated with an increased change in antibody titre against that antigen

following treatment.

164

3.4.1.2 Relationship of pocket depths and antibody titres

An analysis was performed to assess whether there was a correlation between change

in pocket depth and change in antibody titre following treatment (table 3.1) and

between pre-treatment pocket depth and pre-treatment antibody titre against the

various micro-organisms and antigens (table 3.1). No strong correlations were found.

Although the results are not shown, analysis was carried out to investigate a

correlation between pre-treatment pocket depths and change in antibody titre against

all of the micro-organisms and antigens. No significant correlations were found.

3.4.1.3 Other Parameters

Changes in antibody titre against the micro-organisms and antigens tested were

analysed according to smoking status. The results indicate that there is no difference

in the change in antibody titres for smokers and non-smokers.

The changes in antibody titre against the micro-organisms and antigens tested when

analysed according to gender were also examined. The results indicated no significant

difference in change of antibody titres following treatment between males and

females.

Change in antibody titre following treatment, for any of the micro-organisms or

antigens tested, was examined against the age of the subjects at the beginning of the

study. Analyses were carried out to assess whether there were any correlations

between age, change in pocket depth and length of treatment. No correlations were

found to be significant.

3.4.2 Results of Part 2 of the study: Adsorptions

165

Table 3.6 The median percentage reduction in antibody titre following

adsorption compared to the sham adsorbed sera

Median Percentage Range Reduction in Antibody Titre

Against P Sera Adsorbed 65.33 50.00 - 82.40 against P Sera Adsorbed 72.42 40.88 - 86.83 against Aa Sera Adsorbed 61.35 34.03 - 84.73 against Bf Sera Adsorbed 75.00 29.56 - 87.56 against Pi

Against Aa Sera Adsorbed 73.72 25.12 - 86.75 against P Sera Adsorbed 75.19 58.75 - 89.96 against Aa Sera Adsorbed 62.53 26.05 - 89.94 against Bf Sera Adsorbed 74.66 21.86 - 85.20 against Pi

Against Bf Sera Adsorbed 47.79 2.05-73.27 against P Sera Adsorbed 52.13 34.00 - 73.43 against Aa Sera Adsorbed 50.85 26.44 - 70.79 against Bf Sera Adsorbed 48.92 13.85 - 72.61 against Pi

Against Pi Sera Adsorbed 60.91 18.85 - 90.92 against Pg Sera Adsorbed 67.16 25.08 - 91.01 against Aa Sera Adsorbed 63.47 19.62 - 85.10 a ainst Bf Sera Adsorbed 79.83 35.00 - 89.95 against Pi

166

Figure 3.12 Plot of the median percentage reduction in antibody titre against P.

gingivalis following adsorption against P. gingivalis, A. actinomycetemcomitans,

B. forsythus and P. intermedia when compared to the titre observed against the

sham adsorbed sera

100

90

80

70

60

50

40 0

130 20

0

10

sham Pg Aa Bf Pi

Micro-organisms that Sera were Adsorbed Against

167

Figure 3.13 Plot of the median percentage reduction in antibody titre against A.

actinomycetemcomitans following adsorption against P. gingivalis, A.

actinomycetemcomitans, B. forsythus and P. intermedia when compared to the

titre observed against the sham adsorbed sera

100

90

80

70

60

50

40

30

20

10

0

sham Pg Aa Bf Pi


168

Figure 3.14 Plot of the median percentage reduction in antibody titre against B.

forsythus following adsorption against P. gingivalis, A. actinomycetemcomitans, B.

forsythus and P. intermedia when compared to the titre observed against the

sham adsorbed sera

100

90

80

70

60

50

40

30

20

10

0

sham Pg Aa Bf Pi


169

Figure 3.15 Plot of the median percentage reduction in antibody titre against P.

intermedia following adsorption against P. gingivalis, A. actinomycetemcomitans,

B. forsythus and P. intermedia when compared to the titre observed against the

sham adsorbed sera

100

90

80

70 aý

60

0 50

40

30

20

10

0

sham Pg Aa Bf Pi

Micro-organisms that sera were Adsorbed Against

170

Table 3.6 shows the median percentage reduction in antibody titre against P.

gingivalis, A. actin oinycetem coin itans, B. forsythus and P. intermedia following

adsorption. The median reduction is given as a percentage of the serum antibody titre

for the sham adsorbed sera. The sham adsorbed sera gives a value of the antibody

titre following the process of adsorption. No antibodies should have been lost due to

specific binding, however, some may have been lost due to non-specific binding to the

plastic wells of the plate. Sham adsorbed sera was designated to be 100%, and the

median percentage reduction values were given as a percentage of this value. The

table also shows the range values, which are wide, showing a lot of variation between

the values.

When titres against P. gingivalis were measured, 65.33% of the antibody titre had

been lost when the sera had been adsorbed against P. gingivalis itself, 72.42% when

adsorbed against A. actinomycetemcomitans, 62.35% against B. forsythus and 75%

against P. intermedia.

Titres against A. actinomycetemcomitans showed that when compared with the sham

adsorbed sera, 75.19% had been adsorbed out against A. actinomycetemcomitans

itself, 73.72% against P. gingivalis, 62.53% against B. forsythus and 74.66% against

P. intermedia.

Titres against B. forsythus showed that when compared with the sham adsorbed sera

50.85% had been adsorbed against B. forsythus, 47.79% against P. gingivalis, 52.13%

against A. actinomycetemcomitans and 48.92% against P. intermedia.

When titres were measured against P. intermedia, 79.83% was adsorbed out against P.

intermedia itself, 60.91% was adsorbed against P. gingivalis, 67.16% against A.

actinomycetemcomitans and 63.47% against B. forsythus.

3.5 Discussion

It is generally accepted that periodontal treatments, including scaling and root planing,

are highly effective in the treatment of periodontal disease (Badersten et al., 1981,

171

Lindhe et al., 1982, Morrison et al., 1980). Numerous reports have attributed the

success of scaling and plaque control with qualitative and quantitative changes in the

microflora associated with periodontal disease (Genco et al., 1969, Socransky, 1970,

Socransky et at., 1982). Further to this, numerous reports have also considered the

effect of treatment on immunological parameters, many of which are discussed below.

The first part of this study set out to examine the effects of periodontal treatment on

the titres of serum antibodies specific for a range of bacteria, and antigen preparations.

Titres were measured pre-and post-treatment in 25 patients with chronic periodontitis.

The results presented here suggest that periodontal treatment does not have an effect

on serum antibody titres against P. gingivalis, A. actinomyceteincomitans, P.

intermedia, B. forsythus, T. denticola, the OMP of P. gingivalis, A.

actinomycetemcomitans, P. intermedia and B. forsythus and the purified antigens;

HSP60 (of P. gingivalis), LPS (of P. gingivalis), leukotoxin (of A.

actinomycetemcomitans) and the I segment of RgpA protease (of P. gingivalis),

within the time period measured in this study.

A study carried out by Murray et al. (1989) examined the humoral immune response

against P. gingivalis in treated and untreated patients with chronic periodontitis. The

results of this study suggested that serum IgG levels were significantly higher in

subjects with chronic periodontal disease that had not received treatment, compared

with subjects that had received treatment. Murray et al., explained their results as

supporting the notion that removal of antigen by subgingival root planing, i. e. those

receiving treatment, results in a decline in serum antibody levels. This suggests that

treatment has an effect on serum antibody titres against periodontal micro-organisms.

Murray et al. (1989), therefore, appear to agree with numerous studies that have been

reported in the literature, suggesting that periodontal treatment leads to a decrease in

antibody titre. The problem with the study by Murray et al. (1989), is that by having 2

subject groups, one treated and one untreated it is difficult to know or say that the

treated group had a lower antibody titre due to treatment, and that they did not have a

lower antibody titre before they received any treatment. From the knowledge that

every individual has their own distinctive immune repertoire, it is clear that the

humoral immune responsiveness of patients will vary in any study. Therefore, it is

172

difficult to conclude anything from the study described above. For this reason in the

present study, only one group of patients were examined, and the antibody titres for

the one group of subjects were taken before and after treatment, to be sure that any

effects seen, were due to treatment alone.

Another parameter that needs to be taken into account when considering the extensive

literature on this theme is the length of treatment. The period of time that occurs

between the first and second samples being taken varies enormously between studies.

The average length of treatment, i. e. the period of time between the 2 samples being

taken in the current study was approximately 5 months, with a range of 2-10 months.

In terms of other studies on this topic this is quite a short period of time.

Aukhil et al. (1988) carried out a longitudinal study on 24 subjects with periodontal

disease. The study was carried out to investigate the effect of treatment on serum

antibody titres against 10 plaque micro-organisms, including P. gingivalis, P.

intermedia and T. denticola. The antibody titres to the micro-organisms were

measured at 5 different time points; pre-treatment, post hygienic phase, post-surgical

phase, maintenance phase -1 year, maintenance phase -2 years. The results of this

study, in general indicated that periodontal treatment leads to a decrease in antibody

titre. However, many of the subjects were reported to be resistant to change in

antibody titres. The time between treatment and occurrence of significant changes in

antibody titres exceeded one year.

Mouton et al. (1987) also carried out a longitudinal study to investigate the effects of

treatment on serum antibody titres against P. gingivalis in patients with chronic

periodontitis. In this study samples were taken at various time intervals between 13 to

33 weeks post-treatment, and then at one year post-treatment. The overall conclusion

was again that there was a decrease in antibody titres following treatment. Although

as above, the year time point showed a very significant decrease in titre, a significant

decrease in titre was seen approximately 5 months after treatment. The decrease was

shown to be progressive. At 5 months after treatment, the results showed the serum

antibody titres were reduced to 55% of the pre-treatment value and at 1 year post-

treatment, the levels were reduced to 41% of the pre-treatment levels. It must be

173

noted, however, that even at a reduction to 41 % of the pre-treatment levels, these titres

were still 6 times higher than those of periodontally healthy subjects tested.

Tolo et al. (1982) also made a report on the antibody titres pre- and post periodontal

treatment. This study investigated the effect of treatment on IgG, IgA and IgM levels

in the serum against 7 periodontopathogens. Except for P. gingivalis, no significant

changes in titres were recorded against any of the pathogens, when the serum titres

were measured 1 year post-treatment. Successful treatment was said to be correlated

with a significant decrease in titre against P. gingivalis.

Sandholm and Tolo (1986), looked at the effect of treatment on serum antibody levels

in 12 aggressive periodontitis patients. The length of the treatment, i. e. the time

between the first and second sample was 16-32 months. As already discussed, it is

possible, that the length of time following treatment, (that is the time variation in

taking the second sample) may be crucial with regards to measuring the effects of

treatment on antibody titres. Therefore, with such a small number of subjects, 16-32

months is a very large time scale on which to be comparing parameters.

The results of the study reported by Sandholm and Tolo (1986), indicated that

minimal alterations were observed in the levels of antibodies to P. gingivalis,

Capnocytophaga ochracea and Eubacterium saburreum, however, a decrease in

antibody titre against A. actinomycetemcomitans was recorded after treatment. This

decrease was only seen in 3 of the 12 patients, admittedly a quarter of the subject

population, but conclusions cannot really be made on the results from 3 subjects.

Further to this, this study reports that 75% of the patients had low or not detectable

levels of IgG antibodies to P. gingivalis and E. saburreum. If most of the patients had

undetectable IgG levels, it is difficult to properly interpret this study, particularly as

IgG is the most prominent immunoglobulin isotype in the serum, and IgA and IgM

levels would always be expected to be extremely low in the serum of an individual

with a chronic infection.

Horibe et al. (1995) compared the serum antibody titres against 7

periodontopathogens before and after treatment. The post-treatment sample was taken

174

6.9 months on average after the first sample. This study reported that titres against P.

gingivalis and P. intermedia were significantly reduced following treatment, whilst

the titre to the remaining 5 bacteria, showed no significant treatment-related change.

It is very difficult to compare longitudinal studies of this kind. Many differences that

cannot be controlled exist between studies, including patients compliance with the

protocol, the timing of the study and length of treatment, as previously discussed.

Further to this, variations in bacteria and antigen preparation could have an effect on

the results seen. For example, examination of formalin-stable epitope recognition

may provide very different results from those of "native" preparations.

In part 1 of the presently reported study antibody titres pre- and post-treatment were

examined against a large panel of targets. The initial studies examined the antibody

titres against 5 formalin fixed putative periodontal pathogens; P. gingivalis, A.

actinomycetemcomitans, P. intermedia, B. forsythus and T. denticola. The polyvalent

immunoglobulin, IgG and IgA titres were measured for these. None of the titres of

immunoglobulin isotypes changed after treatment. For P. gingivalis, A.

actinonycetemcomitans and P. intermedia there was no significant difference in the

polyvalent immunoglobulin titres and the titres of IgG. For B. forsythus and T.

denticola the IgG titres were slightly lower than the polyvalent immunoglobulin titres.

The titres of IgA against all of the micro-organisms were very low, just above the

background measurement. The low levels of IgA in the serum is not surprising given

the mucosal nature and "local" arm of the IgA response. The titre of IgA was so low

that no confident conclusions could be drawn, however, the analyses that were carried

out indicated no change in titre post-treatment.

As the levels of IgA were so low, a large proportion of the antibodies measured

against the antigenic targets were obviously of the IgG isotype. However, as

mentioned above, the titres of IgG for most of the micro-organisms were not

significantly different from the polyvalent immunoglobulin titres, or on a number of

occasions were slightly lower. Therefore, polyvalent immunoglobulin titres were

measured in this study, to ensure that any changes, which may be significant were

caught.

175

The overall results suggest that there was no significant change in antibody titre after

periodontal treatment, and therefore, do not agree with much of the literature

discussed above. In many of the studies a change in titre was only found to be against

I or 2 of the micro-organisms tested and the antibodies to the rest of the targets tested

remained the same post-treatment. Although the overall results are of "no change", a

few results obtained from this study were interesting and merit discussion.

Table 3.3 shows the correlation between change in pocket depth and change in

antibody titre. There was a positive correlation when change in pocket depth was

correlated with change in antibody titre against B. forsythus whole cells. The

correlation co-efficient was 0.566 (p=0.003) indicating that the correlation is

significant. The change in pocket depth following treatment is always negative i. e. a

decrease in pocket depth, if the treatment is successful. Therefore, these results

suggest that a relationship may exist between the success of treatment and the change

in antibody titre against B. forsythus.

A number of studies have previously suggested that successful periodontal treatment,

either hygiene phase therapy or surgery, are accompanied by a decrease in the

frequency of microbiological detection of levels of B. forsythus (Haffajee et al., 1995;

Haffajee et al., 1997; Socransky and Haffajee, 1993). In terms of previous studies

regarding the effects of periodontal treatment on the immunological parameters of

individuals with disease, there are very few reports. Studies of the humoral immune

response to Gram negative organisms have tended to concentrate on P. gingivalis and

A. actinomycetemcomitans. For this reason it is very difficult to hypothesise about a

role for B. forsythus in chronic periodontal disease. However, the results of this study

suggest that perhaps further studies into the effects of treatment on the humoral

immune response against B. forsythus, maybe of interest.

An interesting result was suggested from the analysis of the correlation between

length of treatment and change in antibody titre to HSP60 (of P. gingivalis). The

correlation appears to be significant and negative, indicating that the shorter the length

of treatment the greater the change in antibody titre. As discussed in the results, it

176

seems that a treatment period of less than 125 days, leads to a positive change in

antibody titre i. e. an increase in antibody titre following treatment, but a treatment

period of 125 days or more leads to a decrease in titre. These results are interesting in

terms of the inoculation theory suggested by Ebersole et al. (1985d). The results of

their study indicated that scaling could lead to a significant increase in circulating

antibodies to various periodontal pathogens and that the peak levels were seen 2 to 4

months following treatment. Therefore, the negative correlation seen between length

of treatment and antibodies to HSP60 is possibly in agreement with the results of

Ebersole et al. (1985d). This result suggests that perhaps we should be concentrating

our attention on the time period of before 125 days after treatment, when an increase

in antibody titre may indicate which pathogens are of importance.

Further studies regarding the length of treatment would be of interest. A longitudinal

study observing the antibody titre of individuals every 2-4 weeks following treatment,

up to 6 months would provide us with more knowledge of the effect of treatment on

the immunological responses. Patient compliance may be difficult to achieve for such

a study.

A large number of predominantly Gram negative bacterial species, possibly 5 to 7 or

more (Schenk, 1985), have been implicated in the aetiology of periodontal disease.

The rationale behind the second part of this study comes from the fact that many

reports have suggested that there are shared antigens amongst the suspected

pathogens. Hence it is of interest to investigate how much of the immune response in

this disease is directed against these antigens. Studies that report an antibody titre

measured against an individual bacterium often do not consider that some of these

antibodies could in fact have been induced against a different bacterium and are cross-

reacting.

A study by Vasel et al. (1996) investigated the cross-reactivity between antibodies

against P. gingivalis and B. forsythus. The approach of their work was different to the

study presented here, however, the results obtained suggest that cross-reactivity does

exist. The study by Vasel et al., investigated this theory using pre- and post-immune

177

serum taken from monkeys following immunisation with P. gingivalis, and

demonstrated a clear cross-reactivity between P. gingivalis and B. forsythus.

A study by Kinane et al. (1993) has also shown cross-reactivity. Antibody titres

against A. actinomycetemcomitans following adsorption against A.

actinomycetemcomitans and H. aphrophilus were measured and the percentage mean

reduction in the titre compared with the sham adsorbed sera. Sera adsorbed against A.

actinomycetemcomitans showed a 68% reduction in titres against A.

actinomycetemcomitans itself and sera adsorbed against H. aphrophilus showed a

similar reduction of 63%. Therefore, most of the reactivity to A.

actinomycetemcomitans could be removed by adsorption with both organisms,

suggesting cross-reactivity between antibodies to these 2 different species.

This study suggests that there is cross-reactivity between P. gingivalis, A.

actinomycetemcomitans, P. intermedia and B. forsythus. The results indicate that in

general the same percentage of antibodies were adsorbed out against each of the

different micro-organisms and this tended to be around 60-80% of the antibodies. The

percentage of antibodies adsorbed out against B. forsythus seemed to be slightly lower

than for the other 3 micro-organisms, around 40-50%. The range values for all of the

results showed a large variation, however, making it difficult to note any real

differences. The large variation seen in the values could have been due to the

different antibody titres in the different patients, thus this fact makes it more difficult

to compare the results as a group.

It was noted that for serum that had been adsorbed against P. gingivalis there was a

higher percentage reduction in antibody titre against P. intermedia and A.

actinomycetemcomitans than P. gingivalis itself. This seems impossible, as there

cannot be a higher percentage of antibodies cross-reactive with P. gingivalis, found in

the patient, than against P. gingivalis itself. Likewise, the same situation was seen

when antibodies were adsorbed against B. forsythus. A higher percentage reduction in

titre was seen against A. actinomycetemcomitans than B. forsythus itself. The reason

for this is not clear, however, as previously mentioned the range of the values is very

wide, and hence there is probably no significant difference between values. The

178

results of this study indicate that the immune response against all 4 of the periodontal

pathogens tested is very similar and there appears to be a significant amount of cross-

reactivity occurring. There appears to be approximately the same percentage of cross-

reactivity between all combinations of the micro-organisms, except for possibly B.

forsythus which appears to have a slightly lower percentage of cross-reacting

antibodies against it.

The data was analysed to see if there were any correlations between the percentage

reduction in antibody titre against the 4 periodontal pathogens and pocket depth of the

patients at the beginning of the study, smoking status and age of the patients. No

correlations were found. In addition, no groupings were noted between either antigens

or in patients.

The apparent involvement of so many microbial agents in periodontal disease makes

identification of the important ones so difficult. In addition, the situation is made

more complex by the variations that exist between the immune repertoires of every

individual. The significance of risk factors and the effects that they have on an

individuals immune response is also unclear.

179

Chapter 4. Immortalised B Cells From Periodontal Disease

Patients.

4.1 Introduction

The discovery, development and exploitation of human monoclonal antibodies has

had a tremendous impact on medical research and now contributes greatly to

diagnostics and therapeutics. Human monoclonal antibody production has permitted

scientists to make huge advances in many areas including: the search for specific and

therapeutic antibodies against human tumours; the production of well tolerated serum

replacements such as anti-rhesus factor D, tetanus, hepatitis immunoglobulin and

other antibodies against infectious diseases; our knowledge of the B cell repertoire in

health and disease giving us significant insights into autoimmunity (Gigliotti et al.,

1984; Van Meel et al., 1985; Stricker et al., 1985; Atlaw et al., 1985; Bron et al.,

1984).

The first description of monoclonal antibody secreting cell lines were reported by

Kohler and Milstein (1975). From this first description, rodent monoclonal

experiments progressed rapidly and have been applied to a wide variety of biological

problems of importance. However, whilst the initial studies to produce human

antibody secreting B cell lines was heralded with much enthusiasm, (Croce et al.,

1980; Olsson and Kaplan, 1980; Steinitz et al., 1977) the available methods carry

many technical problems that remain unsolved. Unfortunately, the generation of

stable human B cell lines which secrete specific monoclonal antibodies is not yet as

routine as the generation of murine monoclonal antibodies.

One aim, as mentioned above, for the production of monoclonal antibodies is to

increase our knowledge of the B cell repertoire and its targets. The discriminating

power of polyvalent serum antibody has limitations. An antigen usually has many

epitopes leading to the production of anti-sera, which is made up of a mixture of

antibodies all with varying specificity's for all the epitopes. Furthermore,

immunisation with an antigen leads to the expansion of various populations of

180

antibody-forming lymphocytes. To observe and investigate these cells is difficult.

These cells can only be maintained in culture for a short period of time, so it is

impossible to grow normal cells and obtain clones that produce antibodies of a single

specificity (Benjamini et al., 1996). Kohler and Milstein (1975) saw an answer to this

problem and it involved the use of a population of malignant plasma cells. Malignant

cells are ìmmortal' and can thus remain in culture for years. They took a malignant

plasma cell population that was deficient in the enzyme hypoxanthine phosphoribosyl

transferase (HPRT). This cell population would die unless they were introduced to

HPRT. They then fused these cells with antibody producing cells that were able to

produce HPRT. The nuclei of the two cell types fused, giving the hybridoma cells the

characteristics of both. Hence, these hybrids were able to manufacture

immunoglobulins as well as survive in culture. The hybrids were grown in a specific

cell culture media, that malignant cells would not survive in without the HPRT

enzyme. Thus allowing the separation of the hybrids from dead malignant cells.

When utilising this method for production, screening is then carried out to identify

hybrids synthesising a specific antibody, which are then cloned.

Cell hybridisation methods are mainly classified into three different categories: (i)

hemagglutinating virus of Japan (HVJ); (ii) Polyethylene glycol (PEG) fusion; and

(iii) electrofusion.

HVJ can agglutinate cells that have HVJ receptors on their cell membrane and cause

efficient cell fusion. HVJ is not, however, really used for preparing hybridomas

because of the possibility that the virus genome may influence hybrid cells. It is also a

very complex procedure. The method was originally reported by Okada et al, (1957)

and showed that animal cells could efficiently be fused using HVJ.

Polyethylene glycol (PEG) is a widely used method because of its simplicity. The

action of PEG is complex, but in brief, it causes localisation of membrane proteins,

allowing direct interaction of lipid bilayers of both cells, making cell fusion easier

(Shirahata et al., 1998).

181

The electrofusion methodology was developed by Zimmerman (1982). Due to

electronic pulses the cell membrane structures are disturbed allowing fusion. The

cells have to be brought into close proximity of each other to be successful, therefore,

to establish close cell-cell adherence, dielectrophoresis of the cells is performed using

sign waves. This causes the cells to align according to the waves and adhere to each

other. Cell fusion is then triggered by the application of electronic pulses.

Electrofusion has an advantage over the popular PEG method in that a much smaller

number of cells are required, often a limiting factor when working with human cells,

however, electrofusion apparatus is expensive and complex (Shirahata et al., 1998).

As mentioned previously, many problems arise in the production of human

monoclonal antibodies. One of the limitations experienced is the choice of lymphoid

tissue for fusion. Animal work tends to use murine tissue and thus the experiments

are not limited by the availability of immune lymphocytes. In murine work, if antigen

substances are available in excess, are non-toxic and sufficiently immunogenic,

immunisation strategies can be optimised to ensure sufficient immune cells are

obtained for fusion (James and Bell, 1987). Obviously, when dealing with human

subjects, the amount and characteristics of the immunogen that can be injected are

much more stringently monitored and there is much more difficulty with safety and

ethical issues.

Another difficulty encountered in human work is deciding what the correct time

period is to harvest the cells following immunisation. This is important as it can

influence their state of differentiation and proliferation (James and Bell, 1987),

however, this is difficult to study due to practical reasons. Studies have indicated that

the optimum time for harvesting peripheral blood cells is 6-7 days (Bogard et al.,

1985) and 3 days post-boosting for murine spleen cells (Schwaber et al., 1984). It

must be noted however, that monoclonal antibodies have been produced using

lymphocytes harvested 1-3 months post boosting (Boyd et al., 1984; Tiebout et al.,

1984; Tiebout et al., 1985; Thompson et al., 1986).

As mentioned previously, human hybridoma production is difficult due to the low

numbers of specific B cells available for fusion, particularly when working with

182

peripheral blood B lymphocytes. In un-immunised patients, the B cells are known to

be in low frequency and often also in a resting state which is incompatible with

successful fusion. It is known that most fusion partners fuse efficiently only with

proliferating B lymphocytes at an undefined stage of differentiation (Johnson and Rao,

1970; Andersson and Melchers, 1978). Therefore, in an attempt to increase the

efficiency of fusion a number of methods have been developed. In vitro immunisation

has been attempted (Borrebaeck, 1989; Borrebaeck et al., 1988), but the rarity of

antigen-specific B cells is still a problem, and is not feasible in the case of molecules

which are poor immunogens.

4.1.2 Epstein-Barr transformation

Human lymphocytes can be immortalised by fusion with a myeloma cell line as

already discussed. They can also be immortalised by a second completely different

method called viral transformation. The virus used in this technique is Epstein-Barr

virus (EBV) and it is usually obtained from the supernatant of the marmoset cell line

B95-8 in culture (Millar and Lipman, 1973). EBV is a herpes virus and the receptor

for EBV on human lymphocytes is the CR2 complement receptor (C3d) (Frade et al.,

1985). Binding of EBV to the CR2 receptor on human B lymphocytes can lead to

insertion on the EBV genome thus causing transformation of the human cells. EBV is

a potent polyclonal activator for human B cells, it is thought that EBV mediated

activation also induces proliferation and clonal expansion of autonomous cell lines

and differentiation into immunoglobulin secreting cells (Rosen et al., 1977; Campling

et al., 1987; Gordon and Graeme, 1988). EBV transformed cells, therefore, can

provide a continual source of immortalised cells for studying B cell-related

phenomena.

EBV can bind to and penetrate all B cells, however, not all B cells are transformed.

Also it has been shown that activation of B cells can be achieved using killed EBV or

by addition of antibody directed against the CR2 receptor, thus activation and

transformation are obviously two different events (James and Bell, 1987). There has

been some confusion regarding the population of cells that are susceptible to

transformation. Arran et al. (1984) suggested that the population of cells that are

183

susceptible to transformation are a small high density resting population that give rise

to IgG and IgA secreting populations. Opposing this view Chan et al. (1986) have

suggested that the population of cells susceptible to transformation is an activated

large cell population that are destined to secrete IgM. Which one of these theories is

correct is unknown, perhaps both are. Many studies suggest a bias towards IgM

secreting cells, agreeing with the theory of Chan et al. (1986), however IgG and IgA is

still found to be secreted by some cell lines. It is thought that only 0.1-5% of

peripheral blood lymphocytes infected with EBV are activated to secrete

immunoglobulin, and this tends to be predominantly of the IgM isotype. Furthermore,

often cells transformed with EBV grow poorly and are low rate immunoglobulin

secretors (Niedbala and Stott, 1998).

Experimental work carried out by Kozbor et al. (1982), combined the techniques of

cell transformation with fusion to obtain hybridomas secreting antibodies against

tetanus toxoid. It is often suggested that fusion of peripheral blood lymphocytes is

unsuccessful because they are not in an actively dividing state. When the cells are not

actively dividing the nuclei of the peripheral blood lymphocytes do not fuse with the

nuclei of the myeloma cells (Burnett et al., 1985). However, following EBV

treatment, even if transformation does not occur, the cells become activated increasing

their chance of successful fusion. The antibodies secreted following fusion are

thought to be representative of the antibodies being produced by the transformed cell

lines at the time of fusion.

4.1.3 The CD40 system

The CD40 system allows expansion of the B cell population obtained, and further

isolation of individual proliferating, differentiating B cell clones (Banchereau et al.,

1991; Banchereau and Rousset, 1991). Stimulation of resting B cells in this system is

an alternative to the EBV method to expand the B cell population before fusion.

CD40 is an integral membrane protein found on the surface of B lymphocytes,

dendritic cells, follicular dendritic cells, hematopoetic progenitor cells, epithelial cells

and carcinomas (Banchereau et al., 1994). The ligand of CD40 (CD40L) is expressed

184

on activated T cells, mostly CD4+ but also some CD8+ as well as basophils and mast

cells. Cross-linking of CD40 with anti-CD40 or CD40L induces strong proliferation

of B cells and addition of IL-4 or IL-13 allows the generation of factor-dependent

long-term normal human B cell lines and the secretion of antibody following isotype

switching (Banchereau et al., 1994).

Banchereau and colleagues (1991) reported that IL-4 and monoclonal antibodies to

human CD40, when presented in a cross-linked fashion by Ltk- cells (transfected

mouse fibroblasts expressing FcyRIUCDw32) (Peltz et al., 1988), permit

establishment of factor-dependent long-term human B cell lines. Therefore, these

factors can exert a direct proliferative effect on B cells when presented appropriately.

The initial studies on this technique reported that within 5 weeks, the initial B cell

population had expanded 150-400 times, and that 20-30% of the resting B cells

stimulated by this procedure can generate B cell clones of 50 to a few hundred cells

after 15 days of culture.

The advantages of using the CD40 system before fusion is that the system can

enhance the number of actively dividing and differentiating B lymphocytes and induce

class switching giving higher fusion efficiency. CD40 is thought to be advantageous

over EBV transformation in that EBV cultures frequently grow poorly, producing

unstable cell lines that secrete low levels of IgM antibodies. The CD40 system is

thought to give higher fusion efficiency and induce class switching to IgG (Niebala

and Stott, 1998). This system does however, suffer from the disadvantage that the

rescued cells are the product of polyclonal activation and are therefore,

unrepresentative of the activated B cell population in vivo, however, this is also the

case for EBV transformed B cells.

As previously discussed in the general introduction of this thesis, there is little doubt

that gingivitis and periodontitis are caused by plaque micro-organisms. The

predominance of the B cells and plasma cells in periodontal lesions, and the often

increased antibody titres seen in the serum directed against putative periodontal

pathogens, underlines the importance of the humoral immune response in the

pathogenesis of these diseases. However, little is known about the specificity of the

185

antibodies raised against the antigens, derived from the bacteria responsible for

disease. The aim of the present study was to investigate in detail the target of the

antibody response in patients with periodontal disease. In order to do this, the task

was to produce human monoclonal antibodies from immortalised B cells isolated from

the granulation tissue of diseased sites and the peripheral blood of patients with

current periodontal disease. These antibodies were then screened using the ELISA

technique against a series of oral pathogens. A number of techniques were employed

to achieve this. Firstly, human peripheral blood B cells were expanded in the CD40

system and then fused with a heteromyeloma cell line by the method of PEG fusion.

Secondly, B cells isolated from granulation tissue, taken from diseased sites in

periodontitis patients were fused directly using PEG.

4.2 Methods

4.2.1 Subjects

The subjects used in this study were patients attending Glasgow University Dental

Hospital. The criteria for inclusion in the study was that they were suffering from

advanced chronic periodontitis. Twenty one millilitres of venous blood were

collected from the anterior cubital region using butterfly needles and 3x 7m1 blue-

capped citrate treated Vacutainer tubes.

The inclusion criteria for the subjects from which granulation tissue samples were

taken, was that they were receiving surgery as part of their periodontal therapy.

Granulation tissue was removed during surgery, which was carried out by an

experienced periodontist, and immediately placed into sterile RPMI medium. The

sample was stored at 4°C until use.

4.2.2 Cell culture

Culture of L-cells.

L cells were cultured in a 75cm2 flask at 37°C, 5% CO2 in complete medium. The

medium was changed every 4 days.

186

Culture of CRL 1668 cells.

Cultures were maintained by the addition or replacement of fresh CRL 1668 culture

medium every 2 to 3 days. The cells were cultured in a 75cm2 flask at 370C, 5% CO2.


Irradiation of the L cells was carried out to arrest cell division. The cells were

incubated at 37°C for 3-4 hours with 1Oµg/ml Mitomycin-C. The flask was then

treated with 10ml of versene to remove adherent cells, and centrifugation was then

carried out at 200 xg for 10 minutes. The cells were then washed 3 times in RPMI

1640.

The cells were then enumerated in white cell counting fluid using a Neubauer

haemocytometer and resuspended at 5xl04cells/ml of complete medium. 3.5ml of

cells suspension per well (1.75xl05cells/well) was then placed into a6 well plate, and

the plate was incubated overnight at 37°C, 5% CO2 to allow cell adherence to the

bottom of the wells.

4.2.4 Lymphocyte preparation

The blood was collected from subjects using 3x 7m1 citrate treated vacutainers

following treatment. 10-15m1 of blood was then layered on top of 10ml of

Histopaque-1077, in a sterile falcon tube, and centrifuged at 400 xg for 30 minutes.

Following centrifugation half of the plasma layer was removed, aliquoted and stored

at -80°C for future testing and the other half was discarded.

The mononuclear cell layer was then transferred using a Pasteur pipette into a clean

sterile flacon tube and PBS was added to increase the volume to 50ml. Further

centrifugation was then carried out at 200 xg for 20 minutes. The supernatant was

then discarded and the pellet was again resuspended in 50ml of PBS, and again

centrifugation was carried out at 200 xg for 10 minutes. Following this the

supernatant was again discarded before the pellet was washed twice under the same

187

centrifugation conditions in RPMI 1640. The pellet was then resuspended in 1 ml of

RPMI and a cell count carried out in nigrosin in 7% acetic acid.

4.2.5 Preparation of 2-aminoethylisothiouronium bromide (AET) -treated sheep red

blood cells (SRBC)

Sheep red blood cells (SRBC) were washed 5 times in sterile saline. This constituted

centrifugation at 200 x g, for 10 minutes and resuspension of the pellet in 20m1 of

sterile saline 5 times. 0.143M AET solution was prepared and filtered, and then I

volume of washed packed SRBC was mixed in a sterile falcon tube with 4 volumes of

AET solution. This mixture was then incubated at 37°C for 15 minutes; however,

mixing at every 5 minute interval was carried out.

The cells were then centrifuged at 200 xg for 10 minutes and washed 5 times in ice

cold saline, and once in RPMI 1640. The cells were then made up to a 10% solution

in RPMI containing 10% FCS. This suspension can be stored for up to 5 days.

4.2.6 T cell depletion

Following cell counting the lymphocytes were prepared at 2x106cells/ml of RPMI

medium. The AET treated SRBC were prepared at a concentration of 0.5%. Equal

volumes were then mixed together and incubated for 15 minutes at 37°C, with mixing

at every 5 minute interval. The suspension was then centrifuged at 200 xg for 10

minutes and then incubated on ice for 30 minutes. The pellet was then resuspended

by gentle shaking in the same supernatant. The suspension was then layered on to

15ml of Histopaque-1077 and centrifuged at 300 xg for 30 minutes. The interface of

B cells was then removed and washed twice in RPMI.

The cells were counted in nigrosin in 7% acetic acid.

188

4.2.7 Expansion of the B cell numbers by the Banchereau method

B cells were added to the adhered L cells at a ratio of 10: 1, therefore, 1.75x10' B cells

were added into each of the wells of the 6 well plate. As the volume of medium in

each of the wells was required to be kept to a minimum and due to further additions

lml of medium was removed from each well before the addition of the B cells. Since

the majority of the L cells should have become adherent to the surface of the well,

they were in little danger of being discarded.

Anti-human CD40 monoclonal antibody was then added at a concentration of 20ng/ml

and human IL-4 at a concentration of 100u/ml. The 6 well plates were then incubated

at 37°C, 5% C02 for 5 days, before the cells were fed using complete medium.

After 11 days the cells were removed, centrifuged at 200 xg for 10 minutes and

counted using nigrosin in 7% acetic acid.

4.2.8 Enzymatic dissociation of granulation tissue

Surgically removed granulation tissue was placed in a sterile 25m1 universal. The

tissue was then cut into 1-2mm3 segments using sterile scissors. The tissue was then

washed twice with RPMI 1640 medium, by centrifugation at 200 xg for 10 minutes.

Cells were then dissociated from the tissue by use of Dispase.

Eight millilitres of Jockliks MEM containing dispase at 3mg/ml (pH 7.4) was added

to the universal containing the tissue. The universal was then incubated at 37°C for 90

minutes with continuous stirring. The medium containing the cells was separated

from the tissue by filtration using 41µm nylon net filters and mixed with excess RMPI

1640 medium.

The cells were harvested by centrifugation at 200 xg for 10 minutes before being

resuspended in Iscoves medium and stored on ice. The cells were then placed on a

mini gradient containing Iml of Histopaque-1077. Centrifugation was then carried

out at 400 xg for 20 minutes, and the interface of cells was collected, washed and

189

resuspended in Iscoves medium. The cells were then enumerated in white cell

counting fluid using a Neubauer haemocytometer. T cell depletion was then carried

out using SRBC as previously described for the peripheral blood lymphocytes.

4.2.9 PEG fusion

Twenty four hours prior to fusion the macrophage feeder layer required for cell

survival was prepared. Macrophages were obtained by peritoneal lavage of a

BALB/C mouse. The cells were counted using white cell counting fluid in RPMI

1640 medium (without phenol red) containing 10% FCS, penicillin, streptomycin and

L-Glutamine at 4x105cells/ml. The cells were then dispensed into 96 well flat

bottomed plates at 50µl/well, ignoring the outside wells of the plate. These plates

were then incubated overnight at 37°C, 5% COz.

The following day the fusion was carried out. The fusion partner cells, CRL- 1668

cells were transferred to a 50m1 centrifuge tube. The lymphocytes were also

transferred into a 50m1 centrifuge tube, and both tubes were centrifuged at 200 xg for

5 minutes. The supernatants were discarded. The cells were then resuspended in Iml

of serum free RPMI 1640 medium. Nine hundred microlitres of nigrosin was

transferred into 2 separate microcaps and one was labelled partner and one

lymphocyte. One hundred microlitres of each of the cell suspensions were then

transferred into these containers, and each sample was counted using a Neubauer

haemocytometer. The volume of each cell suspension was then adjusted with serum

free RPMI 1640 medium to give a density of 106 cells/ml.

Equal volumes of each cell suspension was then transferred to a sterile 50m1

centrifuge tube, and this was topped up with serum free RPMI 1640. Centrifugation

was then carried out at 200 xg for 5 minutes and the supernatant discarded. The tube

was then placed into a small water bath at 37°C. Using a1 ml pipette, I ml of PEG was

added to the cell pellet over a 1-minute time interval, whilst mixing the PEG into the

cells with the pipette tip. The cell suspension was mixed for a further minute in the

water. Again using a1 m] pipette, 1 m] of serum free RPMI 1640 was added to the cell

suspension, again over a minute.

190

Twenty millilitres of serum free RPMI 1640 was then added to the cell suspension

over 5 minutes, starting slowly, dropwise at first then faster as the PEG was diluted

out. The cells were then centrifuged at 200 xg for 5 minutes and the supernatant

discarded. The cells were then resuspended in 30-50ml of HAT medium, and then

cells dispensed into the 96 well flat-bottomed plates that contained the mouse

peritoneal macrophages, that had been prepared the previous day. The wells were

topped up with HAT medium.

The plates were then incubated at 37°C and 5% CO2 for 6 days undisturbed. On day-6

the plates were examined under a microscope to note any wells that had hybrid

growth. The next day, half of the supernatant from the wells was aspirated and the

wells were topped up with HT medium.

The HT medium was then changed every 2-3 days for 7 days, and following this the

cells were cultured in culture medium. When sufficient growth was apparent, the

supernatant was screened using the ELISA technique to investigate the production of

antibodies by the hybrids specific for P. gingivalis, A. actinonryceterncomitans, B.

forsythus, P. intermedia and T. denticola, whole fixed bacteria.

Growth was correlated with antibody activity, and the cells were removed from any

wells in which the surface was covered. Initially these cells were split among 2-4

wells in a fresh plate, then the growth was monitored and the cells were expanded into

8 wells of a 96 well plate, then 4 wells of a 24 well plate, then a 25cm2 flask and then

a 75cm2 flask.

The cell supernatants were constantly checked to make sure that antibody secretion

was maintained.

4.2.10 Growth and preparation of bacteria


and A. actinomycetemcomitans ATCC 29523 (serotype A) were grown and fixed as

191

described in 3.2.2. T. denticola ATCC 35405 were cultured in medium OMIZ-pat +

I% human serum, and were a gift from Dr. C. Wyss, Institut fur Orale Mikrobiologie,

Zurich, Switzerland (Wyss et al., 1997).

4.2.11 Analysis of hybridoma supernatants by the Enzyme Linked Immunsorbent

Assay (ELISA)

Analysis of the antibody production by the hybridomas was analysed by ELISA as

described in 3.2.3. Instead of the addition of sera, the hybridoma supernatant was

added to the wells at this step.

4.3 Results

4.3.1 L-cell culture

The growth of the L cells did not lead to any difficulties at all. The cells grew

extremely quickly and were subcultured every 3-4 days. The cells were cultured in

complete RPMI medium in 5% CO2 at 37°C.

4.3.2 CRL-1668 cell culture

The heteromyeloma cell line CRL-1668 ATCC was much more difficult to culture.

The cells were extremely sensitive to any fluctuations in temperature or CO2 level.

These cells were also very sensitive to cell density and had to be maintained very

strictly between 4x105 and 106 cells per ml of culture medium. This cell line was

cultured in Iscoves medium containing FCS (20%), L-glutamine (4mM) and G418

(200µg/ml) to stabilise its human chromosome complement.


In these studies we activated peripheral blood lymphocytes taken from patients with

chronic periodontitis in the CD40 system. Stimulation of peripheral blood

lymphocytes by the CD40 system dramatically increased the number of actively

192

dividing and differentiating B lymphocytes. The cells grew in tight clusters and the

clusters were observed every 2 days. The numbers of cells increased constantly until

days 10-11, when the cells were counted and then fused with the heteromyeloma cell

line.

Figure 4.1 shows a culture of the L cells and figure 4.2 shows a culture of the CRL-

1668 cells. Figure 4.3 and 4.4 show cultures of B cells in the CD40 system.

193

Figure 4.1

a$ Si .

Figure 4.1 shows L cells in culture. (Magnification x400)

Figure 4.2

Figure 4.2 shows CRL-1668 heteromyeloma cells in culture. (Magnification x400)

Figure 4.3

Figure 4.3 shows cells being cultured in the CD40 system. This photograph was taken on day 3 of culture. A cluster of B cells can be seen proliferating in the centre of the field. (Magnification x400)

Figure 4.4

:. i.

'1 L/

ý41

L1Tý

- ý.

+

4t i5 ý'. e

ý. y

hY ' p

,

"ýY K

l Arý F

Aff) f,

Figure 4.4 shows cells being cultured in the CD40 system. This photograph was taken on day 11. Numerous large clusters of B cells can be seen. (Magnification x400)

4.3.4 Cell fusion

After cell fusion, hybridoma formation was successful. The hybrids were visible by

the eye as growing colonies by about day 10 following fusion. In wells that contained

growing colonies, cloning was carried out by the limited dilution method. When the

cloned cells had grown to visible colony size again the supernatant was removed from

each well and screened for the presence of specific antibodies against periodontal

pathogens. Antibodies specific to P. gingivalis, A. actinomycetemcomitans, P.

intermedia, B. forsythus and T. denticola were screened for.

The screening stage was reached for peripheral blood lymphocytes of 5 patients

expanded in the CD40 system and then fused using PEG, and for directly fused B

lymphocytes isolated from granulation tissue for 6 patients. In all of the II

experiments antibodies specific for the 5 pathogens were identified, and the clusters of

cells were cloned. However, antibody production failed to continue and was not

detected after 3-5 weeks of culture in all of the cases. Therefore, in all of the

experiments, although fusion was achieved, and cloning was successful, the

hybridomas stopped producing antibodies or had died after a period of a few weeks.

4.4 Discussion

The results obtained in this study are disappointing and unfortunately continuation of

this project was not justified due to the lack of success following on from the

laborious work that was carried out in this study. Although, no monoclonal antibodies

were obtained, I feel that the rationale behind the project and the aims of the study

were important to discuss and thus should be included in this thesis.

Two main techniques were employed and two different sources of lymphocytes were

used. The CD40 system was used to expand the number of lymphocytes that were

obtained from the peripheral blood of periodontitis patients. By following the

methodology described by Niedbala and Stott (1998) the CD40 system was optimised

for antibody production, culture conditions and the concentration of IL-4 and anti-

CD40 antibody used. Cells that had been cultured with IL-4 and anti-CD40 presented

196

on Ltk- cells, were viable and grew in tight clusters. The cultures were kept for 10-11

days and provided regular enrichment of the culture medium. This resulted in a 3-5-

fold increase of input B cells within the culture period of 10-11 days. Following

expansion of the B lymphocyte numbers, cell fusion was carried out with the

heteromyeloma cell line.

The second method used in this study involved direct fusion between B lymphocytes

isolated from granulation tissue samples taken from patients with chronic

periodontitis and the heteromyeloma cell line CRL 1668.

The first method involving amplification of peripheral blood lymphocytes in the

CD40 system, followed by fusion, was carried out on 5 patients as previously

mentioned. However, it was then decided to move the study towards direct fusion of

B lymphocytes from diseased tissue. Although a small number of human

heteromyelomas have been shown to be relatively stable, not all B cells fuse equally

well. Schwaber et al. (1984) suggested that the differentiation state of the B cell

seems to affect its ability to fuse. Glassy (1993) reported that small unstimulated

memory cells are resistant to fusion, while larger activated B cells fuse more readily.

Granulation tissue from individuals with periodontal disease is thought to contain

large numbers of B lymphocytes and plasma cells which are thought to comprise a

significant portion of the local cellular infiltrate (Seymour et al., 1979). These B

lymphocytes are thought to be large activated cells. In addition, since the frequency of

antigen-specific B cells in human peripheral blood is very low, the number of

activated B lymphocytes specific to periodontal pathogens present in the tissues would

be expected to be higher.

Peripheral blood B lymphocytes are much more easily obtained from diseased patients

and one blood sample can provide a large number. In addition, the number of B

lymphocytes taken from the blood, can be amplified with use of the CD40 system, as

in this system human resting B cells enter a state of sustained proliferation resulting in

clonal expansion. Therefore, peripheral blood lymphocytes are the best source for

practical and ethical reasons. However, peripheral blood lymphocytes do not make

the best fusion partners. In the mouse it has been suggested that lymphocytes isolated

197

from tissues make better fusion partners than those from peripheral blood. Further to

this it has been proposed that peripheral blood does not contain enough antigen

reactive B cells in the appropriate state of differentiation and proliferation, and

contains too many suppressor and cytotoxic cells and not enough antigen presenting

and T helper cells (James and Bell, 1987).

A study carried out by celenligil and Ebersole (1997) to examine responses to A.

actinomycetemcomitans in patients with chronic periodontitis compared to health,

involved the examination of B cells that had been immortalised using EBV. The

rationale for the study was to investigate whether immortalised B cells could

potentially be used to study the host humoral immune response in this disease. The

results of this study suggested B cells from both diseased patients and controls, were

capable of being immortalised and that EBV transformation led to the proliferation

and differentiation of B cells to give IgM, IgG and IgA secreting cell lines. An

elevated frequency of IgM producing cells was reported and in addition, a higher

proportion of immunoglobulin producing cells were found in the periodontitis patients

compared with the normal controls. Thus celenligil and Ebersole suggested that

either periodontitis patients demonstrate a higher proportion of immunoglobulin

producing cells, or that the circulating B cell pool in periodontitis patients is enriched

for the subset that are EBV transformable. The study went on to investigate the

effects of polyclonal B cell activators, including LPS of A. actinomycetemcomitans on

antibody production by EBV transformed cells.

This study is one of the first in the periodontal disease research area involving human

cell transformation. EBV transformation is a useful tool for providing a resource of

cells for studying human B lymphocytes in health and disease. The elevated

frequency of IgM producing cells noted in this study is not surprising as many studies

in this field have reported a bias in IgM secreting cells. EBV transformation can be

used either alone, as in the study by celenligil and Ebersole (1997), or combined with

cell fusion.

Plasma cells within the gingival tissues have been shown to produce antibody that is

specific for micro-organisms in the subgingival plaque (Ebersole, 1990; Ebersole and

198

Taubman, 1994). Local humoral immune responses are implicated in the

pathogenesis of periodontitis (Pulver et al., 1978; Stem et al., 1981; Torabinejad et

al., 1981; Matthews and Mason, 1983; Smith et al., 1987) and B cells which are

predominant in the lesions locally produce IgG, lower amounts of IgA and small

quantities of IgM, as detected by immunofluorescence techniques (Pulver et al., 1978;

Stern et al., 1981; Torabinejad et al., 1981).

Studies have shown that infiltrating lymphocytes are more likely to arise through

selective homing to the diseased tissues than by local cell division (Takahashi et al.,

1996). In addition, this study suggested that the plasma cells in the tissue samples

observed were among the most active secretary cells in the gingiva. Therefore, it is

thought that antibody-bearing cells specific for periodontal pathogens home to the site

of infection, stay there if they meet relevant antigen or other stimuli, and once resident

remain active, secreting antibodies in a response against the offending pathogenic

bacteria.

Therefore, in theory to study the target of antibody-bearing cells that are present in the

diseased tissue of individuals with periodontal disease, would appear to be a direct

way of examining what has induced the immune response to begin with. The aim of

the project was to fuse antibody-bearing cells from diseased periodontal tissue with a

heteromyeloma fusion partner. A murine fusion partner was not chosen as the aim of

the project was to produce human monoclonal antibodies that could be characterised

fully. A few very satisfactory murine partner cell lines have been developed however,

this is not the case for human partner lines. The fusion rates achieved when using

mouse myelomas as partners are significantly higher than those achieved with human

partners in comparative experiments (Cote et al., 1983; 1984; 1986). In this study a

heteromyeloma cell line was used as a partner for the fusions. This was chosen as it

was more humanised than a mouse partner cell line, however, when using

heteromyeloma cell lines there is always the chance that the heterohybrids may reject

human chromosomes. Thompson et al. (1986) did report, however, that the loss of

antibody secreting lines due to genetic instability is no worse in heterohybrids than in

murine hybrids, hence it was decided that a heteromyeloma was a good choice of

fusion partner.

199

Fusion is complex and extremely technique sensitive and reports have suggested that

the success of the technique can depend on minor technical details such as a particular

batch of PEG, the pH of the PEG when it is in contact with the cells and the length of

time that the cells stay in contact with the fusion partner (Lane et al., 1984; Davidson

and Gerald, 1976). In addition, it is generally accepted that many technical problems

still remain to be solved in terms of successfully fused cell lines that grow well for I

to 2 months before their antibody titre suddenly declines (James and Bell, 1987). In

the present study, it appeared that fusion was successful, however, the cell line did

cease antibody production before cloning could be carried out.

In spite of the numerous difficulties encountered in monoclonal antibody production,

this technology has huge potential. Various human monoclonal antibodies have been

produced to date, many of which have been developed with prophylaxis and therapy

in mind and are, therefore, directed against a wide range of bacterial, viral, erythrocyte

and tumour antigens. The technique has great potential which will only expand and

may eventually provide us with valuable tools for future diagnostics, therapy and

research.

200

Chapter 5. Analysis of the Target of Antibody Secreting

Cells From Diseased Tissue of Chronic Periodontitis Patients

By ELISPOT

5.1 Introduction

Development of the ELISPOT assay was originally based upon the haemolytic plaque

assay developed by Jerne and Nordin (1963). The development of the haemolytic

plaque assay began a new era of detection that permitted antibody analyses to be

undertaken at the cellular level. It allowed cellular immunoglobulin secretion to be

assessed as well as analysis of the regulatory mechanisms that control the expansion

of antibody-producing cell populations. The assay was originally developed to detect

cells secreting complement-binding antibodies against erythrocytes. The use of

haemolysis as a signal in this method necessitates the application of complement and

red blood cells. Both of these components limit the selection of antigens and

antibodies that can be used or assayed. Experimental failures and inconsistent results

are generally related to the difficulties experienced when trying to efficiently couple

antigens to red blood cells (Golub et al., 1968; Pasanen and Mäkelä, 1969; Jerne et

al., 1974).

The ELISPOT technique was developed simultaneously by two different groups;

Czerkinsky et al. (1983) first coined the term ELISPOT and immunohistochemically

showed a spot-forming cell, and Sedgwick and Holt (1983) also reported development

of the assay, but called it the ELISA-plaque. Since the development of the assay,

other names have described the method, such as spot-ELISA and ELISA-spot,

however, the generally accepted name is the ELISPOT. The development of the

technique combined the principles of the ELISA and the plaque forming assay.

However, unlike the ELISA technique, where the final quantification is done by a

spectrophotometer at a defined wavelength, ELISPOT is usually counted manually

using a stereo-microscope. The technique can be very time-consuming and laborious

if there are large numbers of spots, and there is a certain amount of subjectiveness

when counting. For these reasons computer-assisted methods have been developed to

201

aid counting (Cui and Chang., 1997; Herr et al., 1997; Vaquerano et al., 1998). The

benefits that ELISPOT has over its precursor, is that adsorption of antigens to

hydrophobic plastic surfaces is much simpler and easier to accomplish than trying to

couple antigens to erythrocytes, which is the main limitation of the haemolytic plaque

assay. Further to this, adsorption of antigens to plastic surfaces results in this step being stable for a much longer period of time.

Since the introduction of the ELISPOT assay it has been utilised to investigate a

number of different cell products. It has become well established for studying

cytokine-secreting cells (Herr et al., 1997; Hagiwara et al., 1996; Karttunen et al., 1995; Merville, 1993) as well as antibody secreting cells in response to infections and

vaccination. A lot of the work reported on ELISPOT has looked at antibody-secreting

cells in infections that have entered the body via the mucosa such as Salmonella and

Campylobacter (Kantele et al., 1988), typhoid fever (Murphy, 1993) and infection

with Shigella (Murphy, 1993; Orr et al., 1992). The ELISPOT technique has also

been utilised in studies examining urinary tract infections (Kantele et al., 1994), otitis

media (Nieminen et al., 1996), gastritis (Sugiyama et al., 1995) and some viral

infections such as rotavirus gastro-enteritis (Kaila et al., 1992), influenza (Mäkelä et

al., 1995) and HIV (Lee et al., 1989; Nesheim et al., 1992).

The ELISPOT technique has also been used to assess the efficiency of vaccines. This

technique allows the response induced by the vaccine to be studied at the cellular

level. Thus the results are not effected by any pre-existing antibodies that may be

present in the serum of the vaccinated subjects. It has been suggested that vaccines

given by a mucosal route may not increase serum antibody titres, whereas effects can

be observed on a cellular level, by identifying antibody-bearing cells (Kantele and

Mäkelä, 1991), therefore, the ELISPOT is a useful tool. The ELISPOT technique has

been used in studies observing antibody secreting cells following many vaccine

strategies including those against tetanus toxoid (Stevens et al., 1979), influenza

(Yarchoan et al., 1981) and hepatitis B (Cupps et al., 1984).

A complex inflammatory and immune response is involved in the progression of

periodontitis. Tissue activity within the diseased periodontium comprises epithelial

202

and connective tissue turnover, and cellular activity that occurs continuously

throughout the disease process which is associated with infiltrating inflammatory cells

(Page and Ammons, 1974; Page and Schroeder, 1976). Previous work has suggested

that B and T cells accumulate in large numbers in the periodontal tissues, and

Takahashi et al. (1997) demonstrated that there are numerous plasma cells in

periodontitis gingiva. They reported that IgG producing plasma cells predominate in

periodontitis gingiva, with IgA and IgM producing cells being present, but in low

numbers. It has been suggested that the humoral immune response is protective in the

pathogenesis of periodontal disease, and if this is the case, the target or targets of the

antibody secreting cells found in periodontal lesions are clearly of interest and may

indicate the specific microbes involved in this disease.

Ogawa et al. (1989b) analysed the isotype and subclass of antibodies produced by

single cells directed against fimbriae and LPS of P. gingivalis isolated from gingival

tissues of patients with periodontal disease. The study was carried out using the

ELISPOT technique. The periodontitis patients were divided into 3 groups: slight;

moderate; and advanced chronic periodontitis. The total number of plasma cells

increased in relation to the severity of disease, and the major antibody isotype detected

in all subjects was IgG followed by IgA. The results of the study indicated that levels

of antibody secreting cells, or spot forming cells, were higher to fimbriae than to LPS,

which were found to be present in lower numbers. These results came as a surprise

since it is generally assumed that LPS in gingival plaque induces polyclonal antigen-

specific responses. The reasons for this is not clear. It is faesible that LPS of P.

gingivalis is not very immunogenic, as it has been shown to differ chemically from

LPS from other Gram negative bacteria (Mansheim et al., 1978, Koga et al., 1984).

The overall results of this study, indicated that high levels of immunoglobulin, most

notably of the IgG isotype and less, but significant, IgA subclasses are produced

locally in diseased sites. The study also indicated that cells can be isolated from

diseased gingival tissues, and used to investigate the target of the antibodies produced

by these cells. This technique clearly permits us to study more closely the local

immune response induced in this disease.

203

A further study was carried out by the same group (Ogawa et al., 1991), to investigate

anti-fimbriae responses in individuals with chronic periodontitis. The study compared

mononuclear cells isolated from diseased and healthy sites of individuals with chronic

periodontitis. Results were correlated with levels, isotypes and subclasses of anti- fimbriae antibodies found in the sera and those produced by peripheral blood

mononuclear cells of the same subjects. Thus the study was a direct comparison of

the local immune response versus the systemic. The results of this study suggested

that serum antibodies to periodontal bacteria are derived from local stimulation of B

cells in inflamed gingiva. Hence, indicating a need to examine the local immune

response in this disease, as it may provide more details about the pathogens and

antigens that are inducing the immune response, that begins locally and leads to

systemic immunity.

Our understanding of the function of the immune response locally within the tissues is

incomplete, for example we are unsure as to whether the plasma cells, found in large

numbers in lesions, in diseased periodontal tissues are producing relevant or totally

non-specific antibodies to periodontal bacteria. Work carried out on gingiva of

periodontitis patients indicates that the infiltrating lymphocytes are more likely to

arise through selective homing, than by local cell division (Takahashi et al., 1996),

hence indicating the plasma cells present in the lesion are there for a specific reason,

probably producing specific antibody.

Many studies have indicated that antibodies directed against putative

periodontopathogens are present in the serum and GCF of individuals with

periodontitis. Some studies suggest that titres are elevated in these patients compared

with healthy individuals. It is generally accepted that systemically antibodies are

being produced against periodontal bacteria and play a role in the host protective

system. Controversy lies, however, in how much the local immune response reflects

the systemic specificity.

The present study is an investigation, utilising the ELISPOT technique, into the target

of antibody secreting plasma cells isolated from granulation tissue taken from

diseased patients. Moscow and Poison (1991) suggested that experiments carried out

204

on superficial gingival biopsies may not be a true reflection of the disease progression

in the deeper granulation tissues. Therefore, the present study is an investigation of

the targets of some of the plasma cells isolated from granulation tissue taken from

patients with advanced chronic periodontitis. The specific aim of this study was to

identify whether antibody-bearing cells removed from these lesions were secreting

antibody specific for the putative periodontopathogens P. gingivalis, A.

actinomycetemcomitans, P. intermedia and B. forsythus.

5.2 Methods

5.2.1 Patients and tissue samples

The subjects included in this study were 5 patients attending Glasgow University

Dental Hospital. The criteria for inclusion was that they had advanced periodontal

disease and were receiving surgery as part of their periodontal therapy. During

surgery, which was carried out by an experienced periodontist, granulation tissue was

removed and immediately placed into sterile RPMI medium. The tissue sample was

then stored at 4°C until use.



and A. actinomycetemcomitans ATCC 29523 (serotype A) were grown as described in

3.2.2.

Bacterial sonicate preparations were prepared as described in section 3.3.2.

The protein concentrations of the bacterial sonicate preparations were determined

using the Pierce BCA protein assay kit.

205

5.2.3 Enzymatic dissociation of granulation tissue

Surgically removed granulation tissue from each subject was placed in a sterile 25m1

universal. The tissue was then cut into 1-2mm3 segments using sterile scissors,

washed twice with RPMI 1640 medium, and then centrifuged at 200 xg for 10

minutes. Cells were then dissociated from the tissue.

8ml of Jockliks MEM containing dispase at 3mg/ml (pH 7.4) was added to the

universal containing the tissue. The universal was then incubated at 37°C for 90

minutes with continuous stirring. The medium containing the cells was separated

from the tissue by filtration using 41µm nylon net filters and mixed with excess RPMI

1640 medium.

The cells were harvested by centrifugation at 200 xg for 10 minutes before being

resuspended in Iscoves medium and stored on ice. The cells were then placed on a

mini gradient containing Iml of Histopaque-1077. Centrifugation was then carried

out at 400 xg for 20 minutes, and the interface of cells was collected, washed and

resuspended in Iscoves medium. The cells were then enumerated in white cell

counting fluid using a Neubauer haemocytometer.

5.2.4 Enumeration of specific immunoglobulin producing cells by ELISPOT

Multiscreen 96 well HA membrane-bottomed plates designed specifically for the

ELISPOT technique were used in this experiment. The wells were coated with l00µ1

of phosphate buffered saline (PBS) containing sonicated bacterial antigens of P.

gingivalis, A. actinomycetemcomitans, P. intermedia and B. forsythus at a

concentration of 10µg/ml. Tetanus toxoid was diluted in PBS and added at a

concentration of 20pg/ml. This was the control. The plates were then incubated

overnight at 4°C. The following day the plates were washed with 200µl/well of PBS

containing 0.1% Tween-20 (PBS-T) for 5 minutes. The plates were emptied by

flicking and then they were washed twice with 200µI/well of PBS. The wells were

blocked with 200µl/well of Iscoves modified Dulbecco's medium supplemented with

206

5% foetal calf serum (FCS), penicillin (10000M) and streptomycin (1000µg/m1) at

room temperature for 1 hour.

One hundred microlitres was then replaced with medium just prior to addition of the

single cell suspension. The 100µl cell suspension obtained from the granulation tissue

digest, described above, was then added to the wells. Two different concentrations of

cells were used throughout the experiment; lx108 and 5x107 cells/ml. The plates were

then incubated at 37°C for 5 hours. Care was taken not to disturb the plates during

this incubation period as it may cause "double image" spots to form.

After 5 hours the plates were emptied by flicking, the wells were then washed 3 times

with 200µl/well with PBS and 3 times with 200µl/well with PBS-T. Biotinylated

anti-human IgG was then diluted 1: 1000 in PBS-T-FCS and 100 µl/well was added to

the plates. The plates were incubated overnight at 4°C. Following the overnight

incubation the plates were washed 4 times with 200µl/well PBS-T and then

100µ1/well of HRP-conjugated avidin-D was added. This was diluted to a final

concentration of 5µg/ml with PBS-T-FCS. The plates were then incubated for 1 hour

at room temperature. The wells were then washed 3 times with 20041/well PBS-T,

followed by 3 times with 200µl/well PBS. 100m] per well of chromogen substrate

was then added and left for 3 to 8 minutes or until red spots could be visualised. The

reaction was stopped by washing the plate with running tap water. The plates were

then blotted dry.

The nitrocellulose discs covering the bottom of the wells were then removed. They

were then photographed and the spots enumerated using a stereo-microscope equipped

with a vertical overhead white light. The results represent the number of spots

counted per well.

5.2.5 Statistical methods

The data was statistically analysed by the Friedman's test using the Minitab statistical

package. This test was chosen as it compares the number of cells detected against

207

different antigens, when observations have been repeated on the same subjects. In

addition, this test makes no assumptions about the distribution of the data. It was not

satisfactory just to reject the hypothesis and conclude that not all the ELISPOT results

were not identical, therefore, a multiple comparison procedure suitable for use after

the Friedman's test was carried out (Daniel, 1978).

The Wilcoxon Signed Rank test was used to compare the median paired differences

between the results obtained when 108 and 5x 107 cells/ml were added. This test was

chosen as it allowed the comparison of the differences (108 minus 1x107 cells/ml) for

each antigen separately whilst still taking into account that the data has been collected

from the same subjects.

5.3 Results

Many different preliminary experiments were carried out in this area. Three of four

different protocols for the ELISPOT technique were carried out and numerous

problems were experienced. Experiments at the beginning were carried out to try and

get the technique working correctly and these involved studies such as enumeration of

the number of peripheral blood lymphocytes bearing antibody to tetanus toxoid in

healthy individuals following a tetanus vaccination. The results presented in this

chapter are the results obtained from the most successful protocol that was examined.

Establishing the technique was extremely time consuming and laborious, however, the

results presented in this study are the beginning of a project that should be continued

to look at a greater number of chronic periodontitis patients.

Figures 5.1 a-5.4b Show examples of ELISPOT results obtained.

208

Figure 5.1a

Figure 5.1a shows spots indicating plasma cells isolated from granulation tissue

bearing antibodies on their surface specific for P. gingivalis. As measured by the

ELISPOT technique. (Magnification x10)

Figure 5.1b

Figure 5.1b. Negative control for P. gingival is. (Magnification xl 0)

Figure 5.2a


bearing antibodies on their surface specific for A. actinomycetemcomitans. As

measured by the ELISPOT technique. (Magnification x10)

Figure 5.2b

Figure 5.2b. Negative control for A. actinomycetemcomitans. (Magnification xl0)

Figure 5.3a


bearing antibodies on their surface specific for P. intermedia. As measured by the

ELISPOT technique. (Magnification x10)

Figure 5.3b

Figure 5.3b. Negative control for P. intermedia. (Magnification x10)

Figure 5.4a


bearing antibodies on their surface specific for B. forsythus. As measured by the

ELISPOT technique. (Magnification xl0)

Figure 5.4b

Figure 5.4b. Negative control for B. forsythus. (Magnification x10)

d w ü 0 t

4.

0

ro

4°

a 0

x

14 w

h rr

E

AO N ý

o

10 10 ec

w

E

"0

(1)

bA

aý

týA . ,., w

sIodSI'Ia 3° -joqN

MN

sIOdSI'II JO joqN

C E

Cd bi)

O

C QU

bJJ

ýJ

Cý

C 1)

n

00 r- kf) MN

Pg - P. gingivalis

Aa - A. actinomycetenicomitans

Bf - B. forsythus

Pi - P. intermedia

control - Tetanus toxoid

Figure 5.5 shows the median number of ELISPOTs counted against the sonicate

preparations of the 4 periodontal bacteria tested and the control. This plot shows the

results of the experiments when B lymphocytes separated from diseased granulation

tissue were added into the wells at a concentration of 1x 108 cells/ml. Figure 5.6

shows the median number of ELISPOTs counted against the 4 bacteria and the

control. This plot shows the results of the experiment when B lymphocytes were

added into the wells at a concentration of 5x107 cells/ml.

Figures 5.5 and 5.6 show very similar patterns. The line on the graphs indicates the

median values. There is a pattern that can be seen that indicates that the highest

number of ELISPOTs is detected against B. forsythus, increasing from P. gingivalis,

A. actinomycetemcomitans and falling again for P. intermedia. The patterns seen are

very similar for both cell concentrations, however, the pattern is suggesting that there

are a lower number of ELISPOTs detected against the antigens when 5x107 cells/ml g are added compared to 10.

214

Table 5.1 To show the median number of ELISPOTs enumerated against the

periodontal bacteria tested

Median

(108)

Interquartile

Range

Range p-value of

Friedman's test*

Pg 66.1 54.8-74.0 48.0-74.5

Aa 73.0 49.0-89.0 39.0-91.5

Pi 73.0 61.5-78.0 55.0- 81.0 0.01

Bf 81.0 72.6-84.8 68.6-88.5

Control 25.5 17.7-31.0 13.0-35.0

Median

(5x107)

Interquartile

Range

Range p-value of

Friedman's test*

Pg 46.50 37.5 - 57.3 29.5-65.0

Aa 53.00 42.0-61.8 40.5-65.0

Pi 52.00 44.5-60.5 39.0-61.0 0.007

Bf 61.70 49.4-70.5 46.8-75.0

Control 21.00 15.2-25.8 10.0-28.0

* p-value for hypothesis of test of no difference in the number of ELISPOTs detected

against different antigens for each cell concentration independently.

215

Table 5.2 To show the median difference in the number of ELISPOTs

enumerated for the two different cell concentrations

Median Difference

(108 - 5x107

cells/ml)

Interquartile Range

of Differences

p-value* for test of

median difference

equal to 0

Pg 16.6 12.3-23.3 0.059

Aa 20.0 7.0-27.3 0.106

Pi 21.0 9.5-25.0 0.059

Bf 10.5 6.5-35.4 0.059

Control 3.0 2.0 - 6.5 0.059

* p-value for hypothesis test of no difference in the number of ELISPOTs detected

against an antigen for two different cell concentrations.

216

Table 5.1 shows the median values for the number of ELISPOTs enumerated against

the four periodontal pathogens and the control for the two different cell counts.

Table 5.2 shows the median difference in the number of ELISPOTs counted for the

two different cell concentrations added.

Statistical analysis of the results using the Friedman's test was carried out for the 2

different concentrations of cells separately. Both showed p-values of < 0.05. This

indicates that there are differences in the numbers of ELISPOTs detected between at

least 2 micro-organisms for both cell concentrations. On observation of the results in

tabular form, it appears that the median number of cells detected for the control

tetanus toxoid is lower than the median number of ELISPOTs detected against any of

the periodontal pathogens, as expected. The graphical representation of the results

suggests that the trend for the number of ELISPOTs detected against the micro-

organisms when 108 cells/ml are added is very similar to the trend when 5x107

cells/ml are added. By observing figures 5.5 and 5.6 it does seem that the number of

cells detected when 5x107 cells/ml are added is lower than when 108 cells/ml are

added. This is suggested by the shift that is seen in the boxplots, i. e. the boxplot for

5x107 cells/ml has shifted down compared with the one for 108 cells/ml.

Further analysis was carried out using a follow-up multiple comparisons procedure

(Daniel, 1978). The results indicated that when 108 cells/ml were added, there is a

significant difference between the number of ELISPOTs detected against A.

actinomycetemcomitans and tetanus toxoid and between B. forsythus and tetanus

toxoid. The analysis also suggested that when 5x107 cells/ml were added there was a

significant difference between the number of ELISPOTs detected against A.

actinomycetemcomitans and tetanus toxoid and between B. forsythus and tetanus

toxoid. These were the only significant differences found.

The results of the Wilcoxon Signed Rank test on the paired differences in the number

of ELISPOTs detected between the 2 different concentrations of B lymphocytes used

indicated that there was a borderline significant difference between the number of

ELISPOTS detected for P. gingivalis, B. forsythus, P. intermedia and tetanus toxoid.

217

There was no significant difference in the number of ELISPOTs detected when 109

and 5x107 cells/ml were added against A. actinomycetemcomitans.

5.4 Discussion

The ELISPOT technique is a cell-based immunoassay, which means that various

inhibitors present in biological fluids, such as serum, are absent and hence cannot

interfere. Further to this, the quantification of this method is based on cell numbers,

and not arbitrary values, which are much more easily understood. Differences

between high and low antibody secretors can also be disregarded with this method.

The ELISPOT was used to examine the number of antibody secreting cells specific to

putative periodontal pathogens. The cells were isolated from granulation tissue taken

from individuals with chronic periodontitis. The number of cells bearing antibody to

P. gingivalis, A. actinomycetemcomitans, P. interniedia and B. forsythus and tetanus

toxoid were enumerated.

Tetanus toxoid was used as a control, to ensure that binding was due to specific

interactions between antibodies on the surface of cells and the antigens. Tetanus

toxoid was chosen after preliminary studies were carried out to ensure that the

technique could be undertaken with this particular antigen preparation. The antibody-

bearing cells being utilised in this study were isolated from granulation tissue taken

from patients with chronic periodontitis. Granulation tissue is highly vascularised

tissue, full of inflammatory infiltrate. This area of tissue is very rich in immune cells.

It has been reported that at least 50% of the cells in chronic periodontitis lesions are

plasma cells (Kinane and Lindhe, 1997). One would suppose that the majority of the

antibody-bearing cells found in the tissues are specific for periodontal micro-

organisms associated with the disease. Therefore, the reason for choosing tetanus

toxoid as the control is because there should be very few antibody-bearing cells

specific for Clostridium tetani in a periodontitis lesion. Blood vessels are present in

the tissues, and since most individuals have immunity to tetanus toxoid, it provides a

useful control in discounting the fact that we may be merely measuring cells from the

circulation and that the interactions are specific to the tissues.

218

The median number of spots against tetanus toxoid enumerated were 25.5 and 21.0 for

108 cells/ml and 5x 107 cells/ml respectively. Whether these spots are formed due to a

small number of cells present in the periodontal tissues specific for or cross-reacting

with tetanus toxoid, or that they represent non-specific background binding, is not

clear. In addition, Clostridium tetani is a Gram negative bacillus, so the formation of

spots due to cross-reactivity of antibodies cannot be disregarded.

The results suggest that there is no significant difference between the number of

antibody-bearing cells in the granulation tissue against the 4 periodontal pathogens

tested. In addition, the statistical analysis indicates that there is only a significant

difference between the control and A. actinomycetemcomitans and B. forsythus. This

is very difficult to accept because when the median values are observed, it does appear

that there is a large difference between the number of ELISPOTs enumerated against

the periodontal pathogens and tetanus toxoid. Further to this, one would certainly

expect there to be a difference between the number of ELISPOTs detected against the

periodontal pathogens and the number detected against the control. The reason for

this is probably the small sample size. The multiple comparison analyses that were

carried out, made 10 pairway comparisons based on 5 people. They also compared

sum of ranks instead of median values. In addition, the data range is wide, which

cannot be corrected for with a sample size of 5.

Therefore, although the statistical analysis does not indicate that there is a significant

difference, it can be seen from table 5.1 that there does appear to be a difference

between the numbers of ELISPOTs detected against the periodontal pathogens and the

control tetanus toxoid. This suggests that a significant proportion of the ELISPOTs

enumerated for the periodontal pathogens which are due to specific binding and are

not non-specific or due to another phenomenon like those detected against tetanus

toxoid.

The statistical analysis indicates that there is a borderline significant difference

between the number of ELISPOTs detected when 108 and 5x107 cells/ml were added.

The results indicate that for P. gingivalis, B. forsythus, P. intermedia and tetanus

219

toxoid, there are significantly more ELISPOTs enumerated when 108 cells/ml are

added compared to when 5x 107 cells/ml are added. It is expected that there would be

a greater number of ELISPOTs detected when a greater number of cells per well are

added. Further evidence for this, as already discussed, is seen by the shift in the

boxplots in figures 5.5 and 5.6, where there is an obvious shift to a decrease in the

number of ELISPOTs enumerated when 5x 107 cells/ml were added compared to 108

cells/ml.

Although statistical analysis of the results is important and various differences were

significant. It must be stressed that in this study positive results were obtained.

ELISPOTs were detected against the 4 periodontal pathogens tested, indicating that

antibody-bearing plasma cells specific to P. gingivalis, A. actinomycetemcomitans, P.

intermedia and B. forsythus are present in the granulation tissue of the patients in this

study. This indicates that there is a local immune response directed against these

periodontal pathogens occurring in the chronic periodontitis lesion of these patients.

The importance of this phenomenon is that the ELISPOT technique could be utilised

to observe antibody specificity in the local immune response of this disease. It could

further be used for studies examining other diseases, where tissue can be removed

ethically and the antibody-bearing cells isolated for investigation. Examining the

target of the local immune response in periodontitis patients may provide us with

knowledge of the specific immune response and its target. Answers to these kind of

enigmas would lead to research into a new era, in terms of vaccine development and

diagnostic tools.

This study utilises the ELISPOT technique to examine the bacterial target of antibody-

bearing cells isolated from the granulation tissue of patients with chronic

periodontitis. The main limitation experienced with the use of the technique was the

need for a high number of cells. Much of the literature describes the use of the

ELISPOT technique in enumeration of antibody-bearing cells following

immunisation. Cox et al. (1994) reported a study that examined the number of

antibody-bearing cells specific for the influenza virus over a time period directly

following immunisation. Eight blood samples were taken during the period up to 22

days following immunisation. The aim was to examine the time course of appearance

220

of influenza antibody secreting cells in the circulation of individuals following

vaccination. The mean number of specific antibody-bearing cells in the circulation

peaked as high as 606 + 283 cells (mean + S. D. ) per 500,000 cells assayed. In the

present study, it was not expected to obtain a sample containing such a high number

of antibody-bearing cells.

Blood samples would not have been useful in this study, as the length of time since

the onset of chronic periodontitis in the patients was unclear, but certainly it was much

greater than the initial primary immune response stage, which would be a matter of

weeks. The number of cells in the circulation bearing antibody specific for

periodontal pathogens would have been low and the chance of getting even one "hit"

would have been extremely small. For this reason granulation tissue samples were

chosen.

Very little information exists regarding the use of the ELISPOT technique in the

periodontal literature. A study by Kusumoto et al. (1993) identified P. gingivalis

fimbriae specific antibody secreting cells in the lymphoid tissues of mice. Hirsch et

al. (1988) utilised the ELISPOT technique to enumerate antibody secreting cells

specific for types I-Vi collagen in periodontal tissues. Ogawa et al. (1989b) analysed

the distribution of IgM, IgG and IgA secreting cells in gingival tissues from subjects

with periodontal disease using the ELISPOT technique. In addition, Ogawa and co-

workers, have also carried out studies utilising the ELISPOT technique to investigate

the number of antibody-bearing cells specific to fimbriae and LPS of P. gingivalis

(Ogawa et al., 1989a; Ogawa et al., 1991).

Although those latter studies were investigating the target of the human response,

most of the studies in the literature utilise the ELISPOT technique for other purposes.

The aim of the current study was to use the technique to investigate the target of the

immune response, specifically to look at the number of antibody-bearing cells specific

to 4 putative periodontopathogens. Many preliminary studies were carried out to

optimise different factors, including experiments to optimise the concentration of

bacteria used for coating and for the number of lymphocytes added. Problems were

221

encountered when whole fixed bacterial preparations were utilised, and background

staining was very high.

Future studies are justified in this area, and would include a similar study with a much

larger subject number. In addition, the technique could be used to investigate the

presence of antibody-bearing cells specific for purified antigens and eventually for

single antigenic epitopes.

222

Chapter 6. Detection of Antibody-Bearing Plasma Cells in

Periodontitis Granulation and Gingival Biopsy Tissue

6.1 Introduction

The central theme of this thesis is the humoral immune response and its important role

in periodontal disease. Numerous investigations have focused on the immunological

parameters of body fluids such a peripheral blood or GCF. However, there is much

less literature on the local immune response at the site of the lesion. Furthermore,

studies on serum antibody levels or characterisation of peripheral blood lymphocytes

in patients, may not reflect the activity of cells present in the periodontal lesion and

gingival tissues.

Histological studies on inflammatory periodontal lesions were first published as early

as the start of the last century (Zamensky, 1902; Hopewell-Smith, 1911). With the

advent of improved microscope technology during the 1920's more detailed

histopathological analysis of these tissues were reported (Box, 1924; James and

Counsel, 1927; Gottlieb, 1927; Becks, 1927). Much work has been carried out since

these early studies, and as laboratory techniques have been developed the knowledge

of the disease histopathology has grown. It may be noted however, that the early

structural observations are still considered accurate by many investigators.

The major pathological features of periodontal diseases are: (i) apical migration of

epithelial attachment; (ii) an accumulation of inflammatory infiltrate within the

periodontal pocket tissues; (iii) breakdown of connective tissue fibres anchoring the

root to alveolar bone; and (iv) resorption of the marginal portion of the alveolar bone

which eventually results in bone loss. The clinical and histopathological stages of

health to periodontal disease were systematically defined originally by Page and

Schroeder (1976), who developed a classification system to categorise the different

stages of the disease process. However, much of this work was carried out on animal

and adolescent biopsies, so for this reason the following discussion will use the more

223

recent system published by Kinane and Lindhe (1997), which is based on that of Page

and Schroeder.

The system reported by Kinane and Lindhe (1997) classifies the disease progression

from health to advanced periodontal disease. The classification system includes: the

pristine gingiva (Figure 1.2), which is histological perfection; the normal healthy

gingiva (Figure 1.3); early gingivitis (Figure 1.4); established gingivitis (Figure 1.5);

and periodontitis (Figure 1.6).

The pathogenesis of the different stages are discussed elsewhere in this thesis. Of

importance to this study is the histopathogenesis of periodontal disease. It is accepted

that gingival tissues removed from individuals suffering from inflammatory

periodontal disease, contain large numbers of lymphocytes and plasma cells, which

are found in additional numbers with increasing severity of periodontal disease

(Zachinsky, 1954; Brill 1960; Zachrisson, 1968; Payne et al., 1975). Much work has

been carried out to enumerate the plasma cells present in diseased tissues from

patients with advanced periodontal disease, as compared with the number of

lymphocytes present. The results of work carried out by different investigators are not

in agreement, and do not present a clear picture.

Lindhe et at. (1973) reported that the cellular infiltrate of diseased tissues is

predominantly made up of plasma cells. Wittwer et al. (1969) and Angelopoulas

(1973) reported that the number of plasma cells was equivalent to or in some cases

was exceeded by, the number of lymphocytes. Platt et al. (1970) reported that there

are approximately 5 times more lymphocytes than plasma cells present. Mackler et al.

(1977) carried out a study to try and define the relative distribution of lymphocytes

and plasma cells by investigating 3 different types of tissues. The tissues were

gingival biopsies taken from areas characterised as clinically normal, mild gingivitis

and periodontitis. Mackler et al., (1977) reported a difference in cellular profile

between the mild gingivitis biopsies and those from periodontitis sites. The cellular

profile of the gingivitis biopsies was characterised by non-immunoglobulin-bearing

lymphocytes and very few plasma cells. Therefore, the non-immunoglobulin-bearing

lymphocytes could either be T lymphocytes or immature B lymphocytes or a

224

combination of both (MacDermott et al., 1975). In contrast to the cellular profile

from the gingivitis biopsies, Mackler et al. (1977) reported that the cellular profile of

the periodontal tissue biopsies was different, and mainly consisted of

immunoglobulin-bearing lymphocytes and plasma cells. The number of

immunoglobulin-bearing lymphocytes exceeded the number of plasma cells.

More recently Liljenberg et at. (1994), compared plasma cell densities in sites with

active progressive periodontitis, and in sites with deep pockets and gingivitis but no

significant attachment loss over a2 year period. The density of plasma cells (51.3%)

was significantly increased in active sites compared to inactive sites (31 %), indicating

that plasma cells are more dominant in the advanced lesion.

Although studies investigating the numbers of immunoglobulin-bearing lymphocytes

and plasma cells in the tissues seem to be numerous, it is important to consider the

function and target of these cells. The products of B cells and plasma cells in the

tissues include antibodies which may be active against components and metabolites of

periodontal pathogens (Brandtzaeg, 1966), which may result in immunopathology and

tissue destruction (Carvel and Carr, 1982; Genco et al., 1974). Further to this,

examination of plasma cells and their targets in the diseased tissues, which are

specific to the infection and have undergone affinity maturation and migration to the

site, may lead us to a clearer idea of the more important pathogens and immune

response inducing antigens.

Several studies using a variety of techniques to investigate local defence mechanisms

and their specificity have been carried out over the last few decades.

Schneider et al. (1966) carried out a study to demonstrate that there is a local defence

mechanism, present in gingival tissues, of adults with marginal gingivitis and

periodontitis against bacteria present in the adjacent sulcus or pocket to the site of the

tissue taken. Firstly, bacterial samples were taken from the sulci of inflamed gingiva

and placed in acridine orange dye. Immediately following this, gingival biopsies were

taken from the same area as the bacterial sample had been taken. Frozen sections of

the biopsies were adhered to glass slides by the warmth of a finger. This avoided the

225

use of tissue adherent that could interfere with the analysis process. The first

investigation carried out was of the reaction of one of the sections from each biopsy

with fluorescein conjugated rabbit anti-human globulin globulin. This was to

demonstrate specific staining of immunoglobulins in these diseased gingival tissues.

Intense fluorescence was seen predominantly in the connective tissue. Staining was

also seen within the cytoplasm of the perivascular plasma cells. The second part of

the study utilised further serial sections from the biopsies. The acridine orange

stained bacterial suspensions, prepared at the beginning of the study, were added to

each section and incubated, before washing and the addition of rabbit anti-human

globulin globulin conjugated to fluorescein isothiocyanate. Some of the sections were

stained firstly with the fluorescein conjugated rabbit anti-human globulin globulin and

then allowed to incubate with the acridine orange stained bacteria. The results

showed that the bacteria reacted with the tissues in areas of high antibody

concentration, indicating specific antibody-antigen reactions. The specificity was

confirmed by the fact that there were also reactions seen between gingival tissues

taken from one individual and the bacteria taken from another. Once the gingival

tissues had been reacted with rabbit anti-human globulin globulin the bacterial

reactions were inhibited as the antigen-antibody sites were blocked. This provided

evidence that the binding of the bacteria to the tissues was specific. In most of the

sections observed, individual plasma cells were also seen to be reacting with bacteria.

The bacteria could be observed as single cells lined up around the periphery and

overlying the cytoplasm of plasma cells. Although, this study showed that there are

specific interactions between antibodies present in the tissues and antigens, the

specificity of the bacteria and bacterial targets was not investigated. Further to this,

the bacterial sample taken from the sulcus may not have contained any of the

"relevant" or putative periodontal pathogens. Many antibodies could be present in the

tissues that have migrated in with the exudative fluid during the inflammatory

reaction. They may not be specific for the bacteria added for the experiment and may

have bound to non-specific moieties present on the surface of the bacteria such as

lectins. Such interactions are known to occur between some strains of E. coli and

murine lymphocytes (Mayer et al., 1978). The fact that bacterial samples taken from

other individuals reacted, and that the staining in the serial sections always seemed to

226

be in exactly the same area, even though the bacterial samples were heterogeneous,

appears to further indicate this.

Periodontal disease is a chronic disorder which is multifactorial, and there are a

number of pathogens, with similar antigenic moieties that play a role in the induction

of the immune response. The gingival tissues and periodontium are anatomically

distinct and whether the immune response involved in periodontal lesions is mucosal,

systemic or both is unclear.

Two years later Schonfeld and Kagan (1982) determined the percentage of plasma

cells in initial and advanced lesions, from sections of gingival tissue biopsies taken

from periodontitis patients, that were directed against A. viscosus, strain ATCC 27044

and T14-V, P. gingivalis and A. actinomycetemcomitans. The methodology used in

this study was as theirs described previously (Schonfeld and Kagan, 1980). The

results suggested an important role for P. gingivalis in the pathology of periodontal

disease. In every tissue section studied, 10 initial lesions and 16 advanced periodontal

lesions, at least 50 plasma cells were located. There was no significant difference in

the binding of A. viscosus or A. actinomycetemcomitans between initial or advanced

lesions, and the binding of these pathogens to specific plasma cells was quite low.

However, the binding of P. gingivalis was seen in all of the advanced lesions, but only

in 40% of the initial lesions, and a much higher number of plasma cells in the

advanced lesions bound P. gingivalis. The low specificity towards the other putative

pathogens tested remained unclear. It could be that the antigens associated with these

particular bacteria are not powerful immunogens. The results obtained from this study

seem to emphasise the importance of P. gingivalis in the pathogenesis of periodontal

disease. Numerous other studies have indicated that patients with chronic

periodontitis possess high titres of serum antibody to P. gingivalis (Slots and

Listgarten, 1988; van Winkelhoff et al., 1988a; Moore et al., 1991; Mouton et al.,

1981), therefore, it seems that the specificity of the systemic immune response is

reflected locally (or vice versa).

The aim of the present study was to determine the specificity and numbers of plasma

cells for bacterial antigens present in gingival and granulation tissue sections taken

227

from individuals with chronic periodontitis. The number of specific plasma cells in

gingival and granulation tissue sections were enumerated for P. gingivalis, A.

actinomycetemcomitans, P. intermedia, B. forsythus and T. denticola. The number of

plasma cells that bound E. coli were also enumerated as a control to investigate the

phenomenon of natural or non-immunologically mediated binding. The experiments

were further carried out with sonicate preparations of P. gingivalis, A.

actinomycetemcomitans, P. intermedia and B. forsythus, and the purified antigens

leukotoxin (of A. actinomycetemcomitans), HSP75 (of P. gingivalis) and the ß

segment of the RgpA protease (of P. gingivalis). The specificity of the interactions

between the bacteria and the plasma cells was demonstrated by control studies where

a blocking step was included which consisted of the addition of a vast excess of fab

fragments of goat anti-human IgG.

The methodology required for this study was challenging to develop because of the

need to counteract and discount any binding through the Fc portions of antibodies

with any other cell types present, and ensure that all binding detected was specifically

with the Fab portions of antibodies. It was novel and different from previous studies

because of the need to use Fab fragments to blockade the specific interactions between

labelled bacteria and the antibody secreting cells. The use of a biotin system, with its

inherent amplification, means that there is no requirement for the use of fluorescence

microscopy, making it a much easier, sensitive and quicker method.

6.2 Methods

6.2.1 Preparation of Fab fragments of goat anti-human IgG

F(ab)2 fragments of goat anti-human IgG were obtained and digested with Papain.

0.5m1 of lmg/ml F(ab')2 fragments were incubated with 5µg of Papain contained in

125µ1 of beaded agarose. The incubation was carried out for 30 minutes at 37°C with

agitation. Centrifugation was then carried out at 6000 rpm for 5 minutes and the

supernatant removed and transferred to a sterile tube, and stored at 4°C until use.

228

6.2.2 Subjects and samples

Granulation and gingival tissue specimens were obtained from 5 patients attending

Glasgow Dental Hospital. The criteria for inclusion in the study was that they were to

be suffering from advanced chronic periodontal disease and were receiving surgery as

part of their periodontal therapy. During surgery, carried out by an experienced

periodontist, granulation and gingival tissue was removed. The tissue was

immediately fixed in 10% buffered formaldehyde at room temperature and

subsequently processed and embedded in paraffin wax. Serial sections, 4µm thick

were prepared on silane-coated glass slides using haematoxylin and eosin staining

before immunohistochemistry was undertaken.

The following 3 sections, 6.2.3,6.2.4 and 6.2.5 are a description of the 3 different

methodologies that were used during the development of the technique. The

beginning of the development process was method 1 and changes were made and the

technique progressed through methods 1 to 2 and eventually to 3. The results were

obtained by method 3, however, a description of the development method protocols

are included for completeness and so that the reader can form a good understanding of

the development of the technique.

6.2.3 Method 1

6.2.3.1 Bacteria preparation.


and A. actinomycetemcomitans ATCC 29523 (serotype A) were grown and fixed as

described in 3.2.2. T denticola ATCC 35405 were cultured in medium OMIZ-pat +

1% human serum, and were a gift from Dr. C. Wyss, Institut fur Orale Mikrobiologie,

Zurich, Switzerland (Wyss et al., 1997).

229

6.2.3.2 Immunohistochemistry

The sections were firstly deparafinised by immersion in Xylene x2, then in 99%

Ethanol x2 and in 95% Ethanol all for 5 minutes each. The sections were then washed

in distilled H2O, and PBS x2, again for 5 minutes each. The sections were then

treated with 3% Hydrogen peroxide (H202) in Methanol for 15 minutes. The sections

were then washed in distilled H2O for 5 minutes twice. The sections were then

microwaved in 1mM EDTA (pH 8) for 5 minutes x3, and then allowed to cool to

room temperature before being washed for 5 minutes in PBS.

The slides were wiped dry and 200µl of blocking serum was then layered onto each

section and incubated for 20 minutes at room temperature. The blocking serum was

then tipped off the sections and 200µl of fixed bacteria at a concentration of 1.5x l 08

bacterial cells/ml diluted in blocking serum was layered onto all but one section and

incubated for 1 hour at room temperature. For counting purposes the bacteria were

stained using nigrosin and counted in a Neubauer chamber under oil immersion. The

sections were then washed twice in PBS-T for 5 minutes followed by a wash for 5

minutes in PBS. The sections were then wiped dry and 200µl of Fab fragments of

goat anti-human IgG was added at a concentration of 0.05mg/mI, to block any

endogenous IgG, to all but one section (a different section from the minus antigen

section) and incubated for 1 hour at room temperature.

A sample of human serum taken from a patient with untreated chronic periodontal

disease which was obtained previously and stored at -20°C, was diluted 1/200 in

blocking serum. The sections were washed twice in PBS-T for 5 minutes and then

once in PBS, again for 5 minutes. 200µl of the diluted human serum was then layered

onto each section and they were incubated at room temperature for 30 minutes. The

sections were then again washed x2 in PBS-T and xl in PBS, for 5 minutes each.

200µl of Biotin labelled Fab fragments of goat anti-human IgG at a concentration of

0.05µg/ml in PBS was then added to all sections and incubated at room temperature

for 30 minutes.

230

A biotin avidin HRP complex mixture was then prepared and allowed to stand at

room temperature for 30 minutes. The sections were then washed x2 in PBS-T for 5

minutes and then xl in PBS, again for 5 minutes. 20Oµl of the Biotin avidin HRP

complex mixture was then placed onto each section, and the slides were incubated at

room temperature for 30 minutes. The sections were the washed x2 in PBS-T for 5

minutes and with PBS for 5 minutes.

Developing solution was then prepared, and 300µl was placed onto each section. The

sections were incubated at room temperature for 2-7 minutes. The reaction was

checked during this period by viewing the sections at low power under the

microscope. When the reaction was complete the sections were washed in tap water

several times and were then counter-stained in Mayers haematoxylin for 10 minutes.

The sections were then washed several times in tap water and mounted in hot

glycergel with a 22mm x 22mm cover slip.

6.2.4 Method 2.

6.2.4.1 Antigen preparation.

The bacteria were cultured and fixed as described in 3.2.2. Human serum taken from

a patient suffering from adult periodontitis was diluted 1/100 in PBS and then taken

and heat treated in 10mM EDTA for 30 minutes at 56°C. The fixed bacteria were

then incubated with the heat treated serum for 30 minutes at 37°C. The incubated

preparation was then micro-centrifuged at 7000 rpm for 3 minutes and then washed in

PBS x2. The preparation was then used at 1/50 dilution in goat serum.


The sections were deparafinised and prepared as described in section 6.2.3.2

The slides were wiped dry and 200µl of blocking serum was layered onto each section

and incubated for 20 minutes at room temperature. The blocking serum was tipped

off and 200µl per section of the antigen preparation added to all of the sections except

231

one, and incubated for 1 hour at room temperature. The sections were again washed

x2 in PBS-T and xl in PBS, for 5 minutes each. 200µl of Biotin labelled Fab

fragments of goat anti-human IgG at a concentration of 0.05µg/ml in PBS was added

to all sections and incubated at room temperature for 30 minutes.

A biotin avidin HRP complex mixture was prepared and allowed to stand at room

temperature for 30 minutes. The sections were then washed x2 in PBS-T for 5

minutes and then xl in PBS, again for 5 minutes. 200µl of the Biotin avidin HRP

complex mixture was placed onto each section, and the slides were incubated at room

temperature for 30 minutes. The sections were washed x2 in PBS-T for 5 minutes and

with PBS for 5 minutes.

Developing solution was then prepared, and 3O0µ1 was placed onto each section. The

sections were incubated at room temperature for 2-7 minutes, and the reaction was



several times and were then counter-stained in Mayers haematoxylin for 10 minutes.

The sections were washed several times in tap water and mounted in hot glycergel

with a 22mm x 22mm cover slip.

6.2.5 Method 3

6.2.5.1 Antigen preparation

(a) Fixed bacteria.

The bacteria were cultured and fixed as described in section 3.2.2.

(b) Sonicates.

P. gingivalis, A. actinomycetemcomitans, P. intermedia and B. forsythus were grown

as described as above in 3.2.2. Sonicated bacterial preparations of these were

prepared as described in 3.3.2.

232

The protein concentrations were determined using the Pierce BCA protein assay kit.

(c) Purified antigens

Leukotoxin of A. actinomycetemcomitans

leukotoxin of A. actinomycetemcomitans was purified as described in section 3.2.2.

The leukotoxin was a kind gift from Dr. J. Korostoff, Leon Levy Research Center for

Oral Biology, University of Pennsylvania, Philadelphia, USA.

The 0 segment of the RgpA protease of P. gingivalis

Domains of the RgpA protease of P. gingivalis was prepared as previously described

in section 3.2.2.

The 3 segment of RgpA which was used in this study was a gift from Professor M.

Curtis.

HSP75 of P. gingivalis

HSP75 (of P. gingivalis) was purified and prepared according to the method described

by Hinode et al. (1996). This antigen was a kind gift from Dr. D. Grenier, Universite

Laval, Quebec, Canada.

6.2.5.2 Biotinylation of bacteria and bacterial sonicates

25µg Biotinamidocaproate N-Hydroxysuccimimide ester (Biotin) was diluted into 1 ml

DMF, which was then further diluted 1: 5 with 100mM phosphate buffer. 5mg of

sonicated bacteria and 3x109 fixed bacterial cells were mixed with 200µ1 of the Biotin

solution and incubated at room temperature for 3-4 hours with agitation. The bacterial

suspension was then centrifuged at 13,000 rpm for 3 minutes, washed x3 with PBS

and stored at 4°C until used.

233

6.2.5.3 Biotinylation of purified antigens

100µg Biotinamidocaproate N-Hydroxysuccimimide ester (Biotin) was added to 20µg

of purified antigen and dissolved in a final volume of lml of sterile H20. Dialysis

tubing was boiled in 10mM EDTA for 10 minutes. This was repeated 3 times. The

dialysis tubing was thoroughly rinsed in distilled H20. lml of biotin and antigen

solution was then placed inside a piece of dialysis tubing that had been tied at both

ends. The tubing was then placed inside a large beaker containing PBS with 10mM

EDTA with constant stirring. This was incubated at 4°C for 12 hours. Following this

the dialysis tubing was then placed into a fresh beaker containing only PBS and

incubated at 4°C for a further 12 hours with constant stirring. This step was repeated

a further 4 times. The contents of the dialysis tubing was removed after all the

necessary incubations had been carried out, and placed into aI ml microcap tube.

Microcentrifugation was then carried out at 13,000 rpm for 10 minutes. The

supernatant was removed and stored at 4°C until use. This method was used for

biotinylation of the antigens leukotoxin (of A. actinomycetemcomitans), HSP75 (of P.

gingivalis) and the ß segment of the protease RgpA (of P. gingivalis).


The sections were deparafinised and prepared as described in section 6.2.3.2.

200µl of blocking serum was then layered onto each section of half the slides and

incubated for 20 minutes at room temperature. Blocking serum containing fab

fragments of goat anti-human IgG at a concentration of 0.05mg/ml was added to the

other half of the slides and these sections were then also incubated at room

temperature for 20 minutes. These slides were to act as negative controls. The

blocking serum was then tipped off and 200µl of biotin labelled antigen was added to

each section. Whole bacteria were added at a concentration of 3x10' cells/ml,

sonicate preparations at 100µg/ml and purified antigens at a concentration of 1 pg/ml.

Incubation was carried out for 1 hour at room temperature. The sections were then

again washed x2 in PBS-T and xl in PBS, for 5 minutes each.

234

A biotin avidin HRP complex mixture was prepared and allowed to stand at room

temperature for 30 minutes. The sections were washed x2 in PBS-T for 5 minutes and

then xl in PBS, again for 5 minutes. 200µl of the Biotin avidin HRP complex

mixture was placed onto each section, and the slides were incubated at room

temperature for 30 minutes. The sections were then washed x2 in PBS-T for 5

minutes and with PBS for 5 minutes.

Developing solution was prepared, and 300µl was placed onto each section. The

sections were then incubated at room temperature for 2-7 minutes. The reaction was



several times and were counter-stained in Mayers haematoxylin for 10 minutes. The

sections were then washed several times in tap water and then mounted in hot

glycergel with a 22mm x 22mm cover slip.

Method 3 was chosen to be used as it was the most successful, and hence is the

method that was used for further experiments which will be reported subsequently.

6.2.6 Data collection and analysis

Sections were examined under a light microscope, and the presence of positively

stained cells was confirmed by the appearance of cell surface brown staining with the

morphology of human leukocytes. Six microscopic fields at x200 magnification were

chosen for counting in each of the 5 sections for each micro-organism, for purified

antigens and for each of the tissue types. Serial sections were chosen as far as

possible for each different antigen used, so comparable fields could be identified and

counted. Once cell counts were complete median and standard deviation values of the

population were calculated.

The data was statistically analysed by the Friedman's test using the Minitab statistical

package. This test was chosen as it is used to compare the number of cells specific for

the different antigens when observations have been repeated on the same subjects. In

235

addition, this test makes no assumptions about the distribution of the data. It was not

satisfactory just to reject the hypothesis and conclude that not all the cell counts were

not identical, therefore, a multiple comparison procedure suitable for use after the

Friedman's test was carried out (Daniel, 1978).

The Wilcoxon Signed Rank test was used to compare the median of the paired

differences of the results for the granulation tissue samples minus the results from the

gingival tissue samples. This test was chosen as it allowed the comparison of the

differences (granulation tissue cell count for a particular antigen - gingival tissue cell

count for the same antigen) for each antigen separately whilst still taking into account

that the data has been collected from the same subjects.

6.3 Results

Figures 6.1-6.6 show examples obtained from experiments investigating the number

of specific antibody-bearing cells against P. gingivalis, A. actinomycetemcomitans, P.

intermedia, B. forsythus, T. denticola and E. coli whole fixed bacterial cells in

granulation tissue sections. Figures 6.7-6.9 shows examples obtained from

experiments investigating the number of specific antibody-bearing cells against

HSP75 (of P. gingivalis), leukotoxin (of A. actinomycetemcomitans) and the (3

segment of RgpA (of P. gingivalis) in granulation tissue sections.

236

Figure 6.1a

-qv

*. ... '

ýc it_ ... 8 +d #z s

a"

Figure 6.1b

Figure 6.1a shows granulation tissue plasma cells bearing antibodies to P. gingivalis.

Figure 6.1b shows the control, inhibition of specific P. gingivalis antibody interactions using Fab' fragments of goat anti-human immunoglobulin.

(Magnification x200)

Figure 6.2a

y

r ý

t,

d% it "' r ->ýe

ý �a,

'F, '8 "-ý

ý,. rý ýý '; e xý't1 .ý

x ca 4, h w 4_. W 0

xs, C. Jot

Figure 6.2b

1E 4. T NA

AW"

ý4.

Figure 6.2a shows granulation tissue plasma cells bearing antibodies to

A. actinomycetemcomitans. Figure 6.2b shows the control, inhibition of specific

A. actinomycetemcomitans antibody interactions using Fab' fragments of goat anti-

human immunoglobulin. (Magnification x200)

Figure 6.3a

-f--. 1-- 4*

"p .' .'

""d: - m

IF ýr

c`R` r1Ai a. +" , a4

Figure 6.3a shows granulation tissue plasma cells bearing antibodies to P. intermedia. Figure 6.3b shows the control, inhibition of specific P. intermedia antibody interactions using Fab' fragments of goat anti-human (Magnification x200)

Figure 6.4a

.itN 1 4y ý9 !; ' .rt ja

mot

'art i+ ,ýj ,<$1 i ikº 1-. `, *r Fý

k, ui' F

Ir'sRli + ýr t' +t" ý'

r+ lw'r4ýý `}'IY, ý +ü+ 'ý ýw`" ý Rig

4 4'ta'' j t% fi'" *

Figure 6.4b

p, . ,.. a' ä ý'

# 'ý ai: vtI j4

r k

a'E

ýr; ý ý

Fý`,, ýý i

g ; ý'. m .. >. ý

Figure 6.4a shows granulation tissue plasma cells bearing antibodies to B. forsythus.

Figure 6.4b shows the control, inhibition of specific B. forsythus antibody interactions

using Fab' fragments of goat anti-human immunoglobulin. (Magnification x200)

Figure 6.3b

immunoglobulin.

Figure 6.5a

0

fi

, yam tF. --., mow ýý " ý'1, gyp '+' ' '".

lot

a5 . ß'

9ki rs,

AW . 4*

.1

44A

*c r* + ßbä

shad- . °ý'` . rºsý

Figure 6.5b

Figure 6.5a shows granulation tissue plasma cells bearing antibodies to T. denticola.

Figure 6.5b shows the control, inhibition of specific T. denticola antibody interactions

using Fab' fragments of goat anti-human immunoglobulin. (Magnification x200)

Figure 6.6a Figure 6.6b

Figure 6.6a shows granulation tissue plasma cells bearing antibodies to E. coll. Figure

6.6b shows the control, inhibition of specific E. coli antibody interactions using Fab'

fragments of goat anti-human immunoglobulin. (Magnification x200)

Figure 6.7a

Figure 6.7a shows granulation tissue plasma cells bearing antibodies to HSP60 (of P.

gingivalis). (Magnification x 200)

Figure 6.7b

SLý v,

Figure 6.7b shows the control, inhibition of specific HSP60 (of P. gingivalis) antibody

interactions using Fab' fragments of goat anti-human immunoglobulin.

(Magnification x 200)

Figure 6.8a

Figure 6.8b

` .;,

Figure 6.8b shows the control, inhibition of specific leukotoxin (of

A. actinomycetemcomitans) antibody interactions using Fab' fragments of goat anti-

human immunoglobulin. (Magnification x200)

Figure 6.9a

of RgpA (of P. gingivalis). (Magnification x200)

Figure 6.9b

(of P. gingivalis) antibody interactions using Fab' fragments of goat anti-human

immunoglobulin. (Magnification x200)

Figure 6.9a shows granulation tissue plasma cells bearing antibodies to the 0 segment

Figure 6.9b shows the control, inhibition of specific the ß segment of RgpA protease

Figure 6.10 Plot of number of specific antibody-bearing cells per field against

different concentrations of whole fixed bacteria used. Magnification x200 for

granulation tissue sections from patients with chronic periodontitis (n=1).

Cells per field

70

60

50 -

40

30

20

10

0 0

l1l -------------

123

Numbers of bacteria (x10-7)

Pg - P. gingivalis

Aa - A. actinomycetemcomitans

Pi - P. intermedia

Bf - B. forsythus

Td -T denticola

Ec - E. coli

- -Aa

Bf

Pg

--f-Pi

--0-- Ec

243

Figure 6.10 shows the number of cells per field against the 3 different concentrations

of whole fixed bacteria. There is an indication of a peak in the number of cells

counted per field, when a concentration of 1x107 bacterial cells/section (1.5x10s

cells/ml) is used. As the concentration of whole fixed bacteria added is further

increased, there is a decrease in the number of antibody-bearing cells detected. The

experiment was therefore, carried out using 1x10' bacterial cells/section. These are

the results of a preliminary experiment that was carried out to identify the optimum

concentration of the biotinylated bacterial cells to use in the study. The aim was to

establish the optimum concentration for each bacteria to identify the concentration

which gives a compromise between low background staining with adequate staining

intensity of the appropriate cells. The experiment was performed in triplicate on one

set of granulation tissue sections.

244

Figure 6.11 Plot of number of specific antibody-bearing cells per field against

different concentrations of purified antigen. Magnification x200 for granulation

tissue sections from patients with chronic periodontitis (n=1).

Cells per field

180 -

160

140

120 ý-"'-ý-"ý- /

00

80 f HSP

60 -- -o- -LTX

40 - r- RgpA

20

0 ý-_- 0.00 0.20 0.40 0.60 0.80 1.00

Antigen concentration (µg/m1)

HSP - HSP75 (of P. gingivalis)

LTX - Leukotoxin (of A. actinonycetemcomitans)

RgpA -ß segment of the protease RgpA (of P. gingivalis)

245

Figure 6.11 shows the number of cells per field against 3 different concentrations of

purified antigens. These results were obtained from a preliminary experiment that

was performed in triplicate on one set of granulation tissue sections. A positive

correlation is evident between the antigen concentration and the number of cells

counted per field. This indicates that the higher the concentration of antigen used, the

greater the number of specific antibody-bearing cells are detected. The highest

number of cells are detected when a concentration of 1µg/ml of antigen is used. The

graph suggests that at this concentration a plateau is beginning to be established and

hence this is the optimal concentration to use in the experiment. The aim of this

preliminary study was to establish the optimum concentration for each antigen to

ascertain the level which gives a compromise between low background staining with

adequate staining intensity of the appropriate cells. The results suggest that the

number of cells per field detected against P. intermedia is significantly higher than the

number detected against B. forsythus in this individual patient. However, when the

results from all of the patients are observed, there is no significant difference for the

numbers of cells detected against P. intermedia and B. forsythus.

246

Table 6.1 Median figures for cells per field bearing antibody to the whole bacterial cells and purified antigens tested. Magnification x200 for granulation

tissue sections taken from patients with chronic periodontitis (n=5).

Median Interquartile

Range

Range p-value of

Friedman's

test*

Pg 81.0 37.5 - 141.0 36.0- 150.0

Aa 21.0 17.0-37.5 16.0-50.0

Pi 93.0 30.0- 121.5 26.0- 130.0

Bf 95.0 26.2 - 123.0 25.5 - 125.0 0.002

Td 16.0 14.0-26.5 12.0-33.0

Ec 9.0 5.5-15.0 3.0- 17.0

Pg HSP75 116.2 56.7- 134.1 11.1 - 134.2

Aa LTX 113.2 65.0- 119.5 20.8- 121.3

Pg RgpA 0 -

Table 6.2 Median figures for cells per field bearing antibody to the whole

bacterial cells and purified antigens tested. Magnification x200 for gingival

tissue sections taken from patients with chronic periodontitis (n=5).

Median Interquartile

Range

Range p-value of

Friedman's

test*

Pg 36.0 18.5-37.7 15.0-39.3

Aa 16.0 10.9-17.0 9.3-18.0

Pi 29.0 24.5-32.8 22.0-33.0

Bf 26.0 20.4-37.7 15.7-49.0 0.006

Td 12.0 11.0- 14.0 11.0- 15.0

Ec 9.0 5.5- 11.9 3.0- 12.5

Pg HSP75 48.2 18.4-67.5 9.8-78.0

Aa LTX 17.0 11.9-43.0 10.0-54.8

Pg RgpA 0 - - -

247

For all of the subjects, the number of cells detected against RgpA was zero, thus there

was no variability in the number of cells detected against RgpA across subjects. For

this reason it was decided that the RgpA data should be omitted from the statistical

analysis to enable suitable comparisons to be made, using the Friedmans's analysis.

Table 6.1 shows the median number of specific antibody-bearing plasma cells

detected in granulation tissue against the 6 whole fixed bacteria tested and the purified

antigens HSP75 (of P. gingivalis), leukotoxin (of A. actinomycetemcomitans) and the

(3 segment of RgpA (of P. gingivalis) in granulation tissue. The results represent the

median of the results obtained from 5 patients. The results of the Friedmans analysis

suggest that for at least one antigen there are a larger number of specific plasma cells

identified than for at least one other antigen (p = 0.002). The number of cells detected

by the addition of E. coli constituted the controls. A small median number (9.0) of

cells was detected, suggesting that there is some non-specific binding or cross-

reactivity occurring. Cross-reactivity is possible between antibodies induced against

periodontal pathogens and specific for an epitope found on other Gram negative

bacteria.

Table 6.2 shows the median number of specific antibody-bearing plasma cells

detected in gingival tissue against the 6 whole fixed bacteria tested and the 3 purified

antigens. The results of the Friedmans analysis suggest that for at least one antigen

there are a larger number of specific plasma cells identified than for at least one other

antigen (p = 0.006). The number of cells detected bearing antibody specific to E. coli

was small (9.0), again suggesting that there is some non-specific binding or cross-

reactivity occurring.

Experimental work was carried out on sonicated bacterial preparations of P.

gingivalis, A. actinomycetemcomitans, P. interniedia, B. forsythus, T denticola and E.

coli. The results showed positively stained antibody-bearing cells being detected,

however, the background staining was very high. It was very difficult to decipher if

staining was due to specific interactions or non-specific binding. Observation of the

stained sections suggested that the sonicate preparations had bound non-specifically to

the tissue sections. The results of these experiments are not included in this chapter as

248

the technique is demonstrated much better with the results of the whole fixed micro-

organisms and the purified antigens.

Further analysis was carried out using a suitable follow up multiple comparisons

procedure (Daniel, 1978). The results of this test suggested that in the granulation

tissue samples there was a significant difference in the number of cells bearing

antibodies specific to P. gingivalis and E. coli. In addition there was also a significant

difference (borderline) between the number of cells detected against B. forsythus and

E. coli, P. intermedia and E. coli, leukotoxin (of A. actinomycetemcomitans) and E.

coli and HSP75 (of P. gingivalis) and E. coli.

Multiple comparison analysis indicated that in the gingival tissue sections there was a

significant difference (borderline) between the number of cells detected against P.

gingivalis and E. coli, P. intermedia and E. coli, B. forsythus and E. coli and HSP75

(of P. gingivalis) and E. coli.

249

Table 6.3 Median differences in the number of cells per field for the micro-

organisms and purified antigens tested in the granulation and gingival tissue

samples. Magnification x200 for granulation and gingival tissue sections taken

from patients with chronic periodontitis (n=5).

Median Difference

(Granulation - Gingival Tissue)

Interquartile Range

of Differences

p-value for test of

Median Difference

equal to 0

Pg 59.0 12.0-103.4 0.05

Aa 8.70 4.25-20.5 0.05

Pi 66.0 -0.8-94.0 0.05

Bf 69.0 0.6-90.7 0.18

Td 5.0 0.0-15.0 0.14

Ec 4.0 -3.7-4.8 0.69

Pg HSP75 75.3 5.1 -96.2 0.11

Aa LTX 99.1 26.2- 101.9 0.11

Pg RgpA - - -

250

Table 6.3 shows the median differences in the number of cells detected for the 2

different tissue types (granulation minus gingival). The Wilcoxon Signed Rank test

was utilised to examine the paired differences in the number of cells detected against

the micro-organisms and antigens tested in the 2 different tissue types. Table 6.3

shows the p-values which suggest that there is a significant difference in the number

of cells detected against P. gingivalis, A. actinomycetemcomitans and P. intermedia

between the granulation and gingival tissue samples. Therefore, a significantly higher

number of cells specific for P. gingivalis, A. actinomycetemcomitans and P.

intermedia were detected in the granulation tissue samples compared with the gingival

tissue samples. There was no statistically significant difference in the number of cells

detected in the granulation and gingival tissue samples against B. forsythus, T

denticola, E. coli, HSP75, leukotoxin and the ß segment of RgpA.

251

Table 6.4 Mean figures for the percentage of the plasma cells per field specific

for the antigens tested. Magnification x200 for granulation and gingival tissue

sections taken from patients with chronic periodontitis (n=5).

Granulation Tissue Gingival Tissue Mean S. D. Mean S. D.

% of leukocytes 52.3 6.6 23.3 3.1 which are plasma cells

% plasma cells 10.0 6.4 8.8 3.1 bearing antibody specific to Pg % plasma cells 3.0 1.9 4.7 1.8 bearing antibody specific to Aa % plasma cells 9.5 6.3 8.8 2.6 bearing antibody specific to Pi % plasma cells 9.3 6.3 8.3 3.7 bearing antibody specific to Bf % plasma cells 2.2 1.2 3.9 1.4 bearing antibody specific to Td % plasma cells 1.7 0.8 3.1 1.8 bearing antibody specific to Ec % plasma cells 11.1 6.1 14.5 10.3 bearing antibody specific to HSP % plasma cells 10.8 5.1 8.6 7.0 bearing antibody specific to LTX % plasma cells 0 0 0 0 bearing antibody specific to RgpA

252

To express the results in a more relevant context the number of plasma cells were

enumerated so that the percentage of leukocytes that were plasma cells could be

calculated. Further to this the percentage of plasma cells that were being detected for

each of the antigens were also determined, in both the granulation and gingival

tissues. This enabled a comparison between the 2 tissue types of the percentage of

cells for an antigen, adjusting for the size of inflammatory infiltrate. Table 6.4 shows

the percentage of the plasma cells specific for each antigen used in the study. The

values seen are the mean values. The counts were obtained by counting 6 fields of

each section at x200 magnification.

As can be seen from the table of results 52.3 + 6.6% (mean + S. D. ) of the infiltrate in

the granulation tissue is made up of plasma cells, whereas in the gingival tissue the

figure is only 23.3 + 3.1% (mean + S. D. ). Therefore, the ratio of plasma cells to other

leukocytes is much higher in the granulation tissue than the gingival tissue. The

number of cells detected to be bearing antibody against P. gingivalis, A.

actinomycetemcomitans and P. intermedia in the granulation tissue was found to be

higher than the number of cells specific to these micro-organisms in the gingival

tissue samples. However, when table 6.4 is observed it can be seen that when the

counts are expressed as a percentage of the number of plasma cells per field, the

percentage of the cells bearing antibodies to these micro-organisms do not appear to

be significantly different in the 2 tissue types. The value for the percentage of plasma

cells bearing antibodies to all of the micro-organisms and purified antigens tested

shows no difference between the granulation and gingival tissue samples.

253

Table 6.5 Mean figures for the percentage of plasma cells per field specific for

the whole bacterial cells and antigens tested for each of the patients individually.

Magnification x200 for granulation and gingival tissue sections taken from

patients with chronic periodontitis (n=5).

Mean figures for the percentage of the plasma cells per field bearing specific

antibody to:

GRANULATION TISSUE

Patient Pg Aa Pi Bf Td Ec HSP LTX

1 12.7 2.0 11.2 11.6 1.1 2.1 12.9 11.3

(±5.6) (±0.8) (±7.2) (+7.1) (+1.2) (±1.9) (±5.2) (+3.7)

2 19.0 6.3 12.4 12.4 2.6 2.8 16.7 15.3

(+6.7) (±1.9) (+7.3) (+5.0) (+1.6) (±1.3) (+49) (+2.9)

3 4.2 2.1 3.2 3.0 1.9 0.9 1.3 2.4

(+1.9) (+0.8) (+1.3) (±1.9) (+0.7) (±0.6) (+0.8) (+1.9)

4 3.7 1.5 3.1 2.5 1.4 1.2 9.6 10.6

(+2.3) (±0.9) (+2.2) (+1.5) (±1.2) (+1.3) (±2.3) (±1.2)

5 10.7 3.1 17.6 16.8 4.1 1.6 15.0 14.3

(±4.1) (±0.5) (±8.3) (±7.0) (±1.3) (±0.7) (±3.6) (±4.5)

GINGIVAL TISSUE

Patient Pg Aa Pi Bf Td Ec HSP LTX

1 14.0 6.4 11.9 14.5 5.1 5.0 28.8 20.0

(+13) (±6.8) (±10) (+6.2) (±5.3) (±5.3) (±26) (+18)

2 8.3 4.4 6.4 6.2 2.8 2.0 11.3 4.2

(+3.0) (±1.8) (±2.3) (+3.4) (±0.9) (+1.1) (+2.8) (±2.0)

3 5.5 3.0 10.8 8.9 3.6 1.5 21.4 10.4

(±6.7) (±3.1) (±4.4) (+5.9) (+4.0) (+2.9) (+15) (±4.3)

4 8.0 3.0 6.3 6.0 2.4 1.9 6.1 3.2

(+1.5) (±2.0) (+1.0) (+2.1) (+0.8) (+0.6) (+3.2) (±1.4)

5 8.4 6.7 8.8 5.8 5.8 5.0 4.8 5.3

(+5.8) (±5.9) (±6.9) (+3.3) (+4.4) (±5.5) (+7.1) (+8.2)

254

Figure 6.12a Pie chart to show the percentage of plasma cells specific for each

antigenic target tested in the granulation tissue sample of patient 1.

Pa

Unknown

Pi

3f

Figure 6.12b Pie chart to show the percentage of plasma cells specific for each

antigenic target tested in the gingival tissue sample of patient 1.

Art

Unknown

a

Pi

Td



Unknown Aa

'i



Pg

Unkr

Bf

Td

tit Td



Pg Aa

,,.

UllRIIV W! 1



Pg

Bf

Td

Unknoi



Pg Aap;

Unknown



Pg

Bf

Td

Uni



Pg

Unknown

Td

Pi



Pg

Unknow

Pi

Bf

A

Table 6.5 shows the mean percentages (+ S. D. ) of the plasma cells specific for each of

the antigen targets for each of the 5 patients individually. The mean values are of 6

fields per section of granulation and gingival tissue sample.

Figures 6.12-6.16 represent these values as pie charts, indicating the percentage of

plasma cells whose antigenic targets are unidentified. As can be seen from the pie

charts the percentage of cells specific for P. gingivalis, P. intermedia and B. forsythus,

all seem to be approximately at the same ratio for each patient i. e. if there is a high

percentage of cells specific for P. gingivalis, there is also a high percentage for P.

intermedia and B. forsythus. Although the statistical analysis indicated that there was

no significant difference between the number of plasma cells detected against any of

the antigenic targets tested, in some individuals there appears to be a higher

percentage of cells specific for P. gingivalis, P. intermedia and B. forsythus compared

to A. actinomycetemcomitans and T. denticola. The percentage of plasma cells whose

target is unidentified varies between patients, however, there does not appear to be a

correlation between the percentage of cells unidentified and the tissue type. In 3 out

of the 5 patients however, the main antigenic target or ""unknown" does not seem to

have been picked up at all in this study.

6.4 Discussion

While it is generally accepted that bacteria are the primary factors in the induction of

the inflammatory state, little is known about the bacterial specificity of lymphocytes

and plasma cells in the diseased gingival and periodontal tissues. The aim of this

study was to investigate the target of these cells found in 2 different tissue types;

granulation and gingival tissue. Of the 3 different methods tested in this study method

3 proved to be the most successful.

Method 1 involved the addition of whole formalin fixed bacteria directly on to the

section, followed by the addition of goat anti-human IgG Fab fragments to block any

endogenous or unbound IgG. A sample of human serum taken from a patient with a

an antibody titre previously determined by ELISA as "high" to the putative

periodontal pathogens utilised in this study was then added to the sections. The

260

rationale being that antibodies in the serum would bind the bacteria that remained

bound to antibodies on the surface of plasma cells in the section. Biotin labelled Fab

fragments of goat anti-human IgG were then added, to detect bound serum antibodies,

before amplification and detection of the signal by the addition of biotin avidin HRP

complex and then developing solution. Biotin labelled Fab fragments were used

instead of whole antibodies in the remote possibility that aggregated biotinylated

antibodies would bind to Fc receptors. One potential problem with this method was

that when the antigen bound antibodies on the surface of the plasma cell, epitopes of

the antigen were engaged for binding to occur. Detection of this antigen-antibody

complex was by the addition of human serum antibodies that also had to bind the

antigen. Not all epitopes of the antigen were available for binding to the serum

antibodies. If the serum antibodies that were added for detection purposes were

specific for the epitopes that were not available, detection would not have occurred.

Further to this, although blocking serum was added at the beginning of the experiment

to hinder any non-specific binding sites, it cannot be guaranteed that all were blocked.

Human serum antibodies could bind these unbound sites through Fc interactions,

possibly leading to a false detection signal.

Method 2 was developed with the aim of reducing the number of steps in the protocol.

In this method a sample of serum taken from a high antibody titre patient with

periodontal disease was heat treated and then incubated with the formalin fixed

bacteria. This allowed the addition of the antigen and the serum antibody to be added

in one step. The problems experienced in the first method were still not solved by this

method, as all of the epitopes on the antigen surface were still not available. Whereas

before they were not all available for the human serum antibodies, in this method they

were not all available to the antibodies on the surface of the plasma cells in the tissue

sections. Again, there was no certainty that there were no interactions between the Fc

portions of the human serum antibodies and other cells in the tissue with Fc receptors.

Method 3 involved direct biotinylation of the antigen, which was then added directly

to the section. Detection was achieved through the addition of biotin avidin HRP

complex mixture. This method was less laborious, time consuming and complicated,

and more accurate than methods I and 2. Furthermore, all of the epitopes on the

261

surface of the antigen were available for binding to specific plasma cells and there was

no addition of a further step that could lead to non-specific Fc interactions. The

experiment allowed specific binding of antibody on the surface of plasma cells to

antigen. The antigen had already been biotinylated, and therefore, amplification and

detection was carried out in 2 simple steps. The one problem experienced was that

although blocking serum was added at the beginning of the experiment, there was a

certain amount of non-specific binding of the antigen to the section. More

background staining, and hence non-specific binding was seen when sonicate

preparations were used. The purified antigens showed very little non-specific binding.

Investigations carried out by Schneider et al. (1966) utilised fluorescein conjugated

rabbit anti-human globulin globulin to demonstrate specific staining of

immunoglobulins in diseased gingival tissue sections. Further to the detection of

immunoglobulins, acridine orange stained bacterial suspensions were added on to the

sections to observe if these bacterial preparations were bound by the detected

immunoglobulins. The methods of Schonfeld and Kagan (1980) utilised bacteria that

were labelled with rhodamine, which is a fluorochrome, and these were added onto

gingival tissue sections taken from periodontitis patients.

Although method 3 is quite similar to the methods described by these workers there is

a difference in the detection method. The methods used by Schneider et al., (1966)

and Schonfeld and Kagan (1980; 1982) utilised fluorescence to detect binding

between antibodies on the surface of plasma cells and antigens, whereas in this study,

biotin was used as the detection agent. In this study the Vectastain Elite ABC kit was

used. The principles of this kit are based on avidin, a 68,000 molecular weight

glycoprotein having an extremely high affinity (1015M-1) for biotin. As the affinity of

this interaction is over 108 times higher than that of antibody for most antigens it is

essentially irreversible. In addition the biotin/avidin system can be effectively

exploited because avidin has 4 binding sites for biotin and most proteins, including

antibodies and enzymes, can be conjugated with several molecules of biotin. The

advantages of using biotin are that it provides a permanent record of the results,

whereas fluorescence is known to fade after approximately 7 days. Also the biotin

avidin HRP complex mixture provides an amplification step which can be magnified

262

further by the addition of avidin labelled anti-avidin. Fluorescence cannot be

amplified, and therefore, provides a weaker signal.

A further difference between this work and those of Schonfeld and Kagan (1980,

1982) and Schneider et al. (1966) is that they used frozen tissue sections whereas

paraffin embedded sections were used in this study. A preliminary experiment was

carried out to examine the immunohistochemistry of a frozen specimen compared

with a matched paraffin embedded section. One piece of granulation tissue was taken

and divided into 2 pieces, one piece was frozen and the other paraffin embedded. This

experiment was carried out in triplicate. The results showed that there was no

significant difference in cell numbers detected in the 2 different types of section.

Therefore, paraffin embedded sections were used in this study, as they have several

advantages over frozen sections. They can be stored for many years and are easier to

cut in thin sections (4µm) allowing the same cells to be apparent on adjacent cut

sections. Thin sections hold an added advantage in that they require shorter

incubation times for optimal staining. In addition, the morphology seen in paraffin

embedded sections is much better than in frozen sections and they are much easier to

obtain, as an archive of paraffin embedded sections can be retained. Paraffin sections

are superior to frozen sections, as long as they are completely de-paraffinised before

immunohistochemistry is performed. Ensuring that all of the antigenic epitopes are

unmasked is important, however, this is not experienced with frozen sections. The

preliminary study indicates that the epitopes of the paraffin embedded sections are

being efficiently unmasked.

Preliminary experiments were carried out to establish the optimum concentration for

each antigen. Various concentrations were attempted to ascertain a concentration

which gives a compromise between low background staining with an adequate

staining intensity of the appropriate cells. For the whole fixed bacteria 3 different

concentrations were used 1x107,2x 107 and 3x 107 bacterial cells/section. Figure 6.10

shows that for P. gingivalis, A. actinomycetemcomitans, and P. intermedia there was a

peak number of antibody-bearing cells detected when the antigen was used at a

concentration of 1x 107 bacterial cells/section. It was therefore, decided that this

concentration would be used for all of the experiments. Three concentrations; 0.2,0.5

263

and 1.0µg/ml were tested for the purified antigens. The concentration of 1.0µg/ml

was chosen to be used, as on average a higher number of antibody-bearing cells were

detected.

The results suggest that the number of antibody-bearing cells detected per field

specific for P. gingivalis were significantly higher than detected in the granulation

tissue for E. coli. The number of cells detected against B. forsythus, P. intermedia,

leukotoxin (of A. actinomycetemcomitans) and HSP75 (of P. gingivalis) were also

found to be significantly higher (borderline) than the number of cells detected against

E. coli. In addition, the results for the gingival tissue samples suggest that the number

of cells specific for P. gingivalis, P. intermedia, B. forsythus and HSP75 (of P.

gingivalis) are significantly higher (borderline) than detected for E. coli.

The number of cells per field found in the gingival tissue sections specific for P.

gingivalis, A. actinomycetemcomitans and P. intermedia are significantly lower than

those found in the granulation tissue section. The numbers of cells detected against

the other periodontal pathogens appeared to be lower, however, this was not justified

statistically. They are not significantly lower, even though on observation of the

median values they appear to be the range values for these counts are large.

E. coli was used in the experiment as a control. E. coli is a Gram negative bacteria,

but not a periodontal pathogen, so was utilised to address the issues of non-specific

binding and cross-reactivity. There was a small number of cells detected against E.

coli which could be due to non-specific binding, however, it seems more likely that it

is due to the binding of E. coli to plasma cells bearing cross-reacting antibodies on

their surface. As discussed in the general introduction of this thesis, many antigens

e. g. LPS and HSP, have homology between species. Although the numbers of cells

detected against E. coli were very small, because of the multiple comparison analysis

which is comparing the cell numbers against 9 different antigenic targets for only 5

subjects, the results do not suggest a significant difference in the numbers of cells

detected against E. coli and all of the periodontal pathogens and antigens, however, it

was the case for a number of those tested. Although, statistically this therefore, does

not suggest full confidence in the technique, when the numbers of cells detected

264

against E. coli are examined they are extremely low. In addition, most of the micro-

organisms and antigens are found to have statistically more specific cells against them

when compared to the number recorded against E. coli. Interestingly, the median

number of cells detected against E. coli is almost the same in the granulation and

gingival tissue sections, further suggesting that these cells are due to non-specific

binding.

In this study, a number of antibody-bearing cells specific for the purified antigens

HSP75 (of P. gingivalis), leukotoxin (of A. actinomycetemcomitans) and the ß

segment of the RgpA protease of (P. gingivalis) were examined. In the granulation

and gingival tissue sections there was a high number of cells detected against HSP75

and leukotoxin. There was no significant difference between the numbers of cells

detected for these 2 different purified antigens. There were no cells detected against

the ß segment of RgpA in either tissue types.

Table 6.4 shows the percentage of the plasma cells in a field that are specific to each

micro-organism and antigen tested. The results suggest that a significantly higher

percentage of the leukocytes in granulation tissue are plasma cells when compared

with gingival tissue. The mean percentage of the infiltrate in granulation tissue that

was made up of plasma cells was 52.3% ± 6.6% (mean + S. D. ) compared to 23.3% +

3.1 % (mean ± S. D. ) for gingival tissue.

Granulation tissue is highly vascularised, full of inflammatory infiltrate and found

adjacent to the bone. Gingival tissue is more superficial, associated with the

periodontal pocket and often referred to as infiltrated connective tissue. Therefore,

due to the nature of the 2 tissue types one would expect a greater percentage of

antibody-bearing cells to be detected in the granulation tissue. This area of tissue is

very rich in immune cells. In periodontitis, it has been reported that 50% of the cells

in the established lesion are plasma cells. In addition several investigators have

shown that the inflammatory cell infiltrate in human chronic periodontal lesions is

predominated by lymphocytes and plasma cells (Lindhe et al., 1980; Mackler et al.,

1977; Page and Schroeder, 1976). The results of this study are in agreement with

these reports.

265

The percentage of the plasma cells specific for all of the whole micro-organisms and

the purified antigens is the same for the granulation and gingival tissue samples.

Thus, the percentage of the total number of plasma cells specific to P. gingivalis in the

granulation tissue is not significantly different from the percentage of the total number

of plasma cells in the gingival tissue samples. Therefore, although when looking at

the mean number of plasma cells per field it appears that there is a significantly higher

number of plasma cells specific to P. gingivalis in granulation tissue samples when

compared to gingival tissue, the percentage of the infiltrate that is specific to P.

gingivalis is not significantly different.

The results of the number of cells specific to HSP75 and leukotoxin suggested cell

numbers, similar to those found against P. gingivalis whole bacteria, in granulation

tissue samples. The high numbers of cells bearing antibodies to leukotoxin does not

correlate with the results of the experiments with the whole fixed bacteria A.

actinomycetemcomitans. There was a high number of cells detected against

leukotoxin in the granulation tissue samples, and a moderate number in the gingival

tissues. However, leukotoxin is an antigen taken from the surface of A.

actinomycetemcomitans, and there were very few cells found to be bearing antibody

specific for this micro-organism. One would expect there would be a correlation

between the number of cells found to be bearing antibody against a purified antigen,

and the whole bacteria that it were taken from. Examination of formalin-stable

epitope recognition may provide different results to examination of "native" antigen

preparations. Important epitopes may be destroyed or changed during the fixing

process, and may no longer recognised by specific antibodies. In addition, leukotoxin

is a much smaller molecule than that of the whole bacteria A. actinomycetemcomitans.

The much higher number of cells detected against leukotoxin perhaps highlights a

problem of steric hindrance that is experienced on addition of the whole bacterial

cells. Addition of purified antigens possibly leads to a more sensitive detection

technique. If this is true, the number of cells detected against the whole bacterial cells

is possibly an under estimation, and there could be a higher percentage of plasma cells

specific for the whole micro-organisms in general.

266

For all of the subjects tested there were zero cells detected against the ß segment of

the RgpA protease (of P. gingivalis). It has previously been reported that the (3

segment is the immunogenic portion of this protease (Curtis et al., 1993). IgG

antibodies specific for this portion of RgpA have been detected in the serum of

patients with periodontal disease, e. g. in chapter 3 of this thesis. However, it has been

reported that the (3 segment of RgpA has a primary sequence similar to that of

adhesins and haemagglutinins of other molecules and micro-organisms (Curtis et al., 1993). Therefore, it should be considered that perhaps the serum antibodies are cross-

reacting with this segment of the P. gingivalis protease as there does not appear to be

any evidence of a local immune response in the diseased tissues of periodontitis

patients against this antigenic target. The lack of detection of antibody-bearing cells

to this antigen could however, be due to there not being many in the tissues, either

way this is an area requiring further research.

Figures 6.12-6.16 show the percentages of plasma cells specific for the whole

bacterial cells P. gingivalis, A. actinomycetemcomitans, P. intermedia, B. forsythus

and T. denticola. The percentages of cells detected against E. coli, HSP75 (of P.

gingivalis) and leukotoxin (of A. actinomycetemcomitans) were not included in the pie

charts. The percentages of cells detected against E. coli were not included as it is

thought that the cells shown to be specific for this micro-organism are possibly cross-

reacting antibody-bearing cells. The percentages of cells detected against leukotoxin

and HSP75 were also not included as some of these cells could possibly have been

counted as being specific for any of the other Gram negative bacteria tested in the case

of HSP75, and against A. actinomycetemcomitans in the case of leukotoxin.

The results obtained from this study do not appear to agree with those reported by

Kagan (1980). Kagan utilised the method developed by Schonfeld and Kagan (1980)

to examine the percentage of plasma cells specific to P. gingivalis in initial/early

lesions compared with the percentage in established/advanced lesions. The gingival

tissue samples cannot really be compared with the early/initial lesion samples, as there

is not enough information in the report to allow this. However, the results from the

established/advanced lesion samples can be compared with the results from the

267

granulation tissue samples. Kagan reported that 31.6% of the plasma cells in the

established/advanced lesions were specific to P. gingivalis, whereas this study

reported a mean percentage of only 9.8%. Kagan suggested that a percentage of

31.6% was high, although this could have been due to the small sample size. The

results indicated that established/advanced lesions were examined for only 3 subjects,

which was small. Further to this, the number of plasma cells counted in total in the

study carried out by Kagan (1980) was extremely low, much lower than the number

reported in this study and also that reported in the study by Mackler et al. (1977).

This study was carried out using granulation and gingival tissue sections, to determine

if the target of the antibody-bearing cells in these different tissues differed or

exhibited any different patterns. Lappin et al. (1999) looked at the relative

proportions of leukocytes in granulation tissues samples in patients with chronic and

aggressive periodontitis, and compared these with numbers obtained from gingival

tissues. The results of this study suggested that B cells in the gingiva and the

granulation tissue are long lived cells, and the B cells from both tissue types behave in

a similar fashion. Therefore, it was of interest to compare the targets of the antibody-

bearing cells from both tissue types.

The results of the experiments that attempted to detect plasma cells bearing antibodies

with sonicated bacterial preparations were not included in this study. The experiments

utilising sonicated bacterial preparations showed much more background staining than

seen when whole micro-organisms and purified antigens are used. These results

suggested that some of the binding may have been non-specific and hence, it was

difficult to differentiate between specific and non-specific binding to the sections.

Sonicated bacterial preparations resulted in the exposure of many molecules that are

not on the surface of an organism. Whether some of these are carbohydrate moeties

that are "sticky" and can non-specifically adhere on to the tissue is unknown.

It is considered that the methodological approach to this study has achieved interesting

results particularly as the study was largely developmental. This is not a standard

technique and hence, many preliminary experiments were carried out in the

developmental stages for each part of the technique, such as optimisation of antigen

268

concentrations, blocking techniques, use of frozen and fixed tissue and compete

labelling of antigens. Unlike conventional immunohistochemical techniques where

cells are detected by the use of antibodies, this approach is aimed at achieving the

opposite. The technique is based on the rationale that by the addition of antigen, the

detection of antibody-bearing cells will be accomplished. The resulting data at this

point is difficult to utilise in a quantitative manner, as one will never be certain that all

of the antibody-bearing cells are being detected. In fact, the number of cells detected

against leukotoxin compared with the number detected against A.

actinomycetemcomitans suggests that all of the cells are not being detected. Further to

this, although the preliminary experiment was carried out to compare paraffin

embedded with frozen sections, one can never be totally certain that all of the epitopes

have been unmasked. Therefore, although the method described in this chapter has

proved successful and is a great achievement in terms of a qualitative study, whether it

can be used quantitatively is more debatable.

Method 3 used in this study, is a newly developed method that could prove useful for

the examination of the target of antibody-bearing cells in diseased tissues from all

areas of the body, and in certain infections this could play a role in diagnosis and

choice of therapy. This is a little more difficult in the polymicrobic situation of

periodontal disease, however, if a larger study were carried out, possibly on

individuals in health, with gingivitis and periodontal disease, more patterns may

emerge on the target of antibodies in the tissues at different stages of disease. Of

course, there are always difficulties in such studies, to gain ethical approval for the

removal of tissues from individuals in health and with gingivitis.

In the study, the results suggested that in both granulation and gingival tissue samples,

a high proportion of the antibody-bearing cells were specific for the putative

periodontal pathogen P. gingivalis. There is a higher load of P. gingivalis associated

with active periodontal lesions, than inactive (Dzink et al., 1988; Walker and Gordon,

1990), and following breakdown after treatment, large proportions of P. gingivalis

have been found at sites, indicating an important role for the species in pathogenesis

of recurrent disease. Therefore, from this study it seems that P. gingivalis is important

in the induction of the humoral immune response at the site of infection. Antibody-

269

bearing cells specific for the other periodontal pathogens tested were also identified.

There was no statistical significance difference found between the numbers of cells detected against P. gingivalis, A. actinomycetemcomitans, P. intermedia, B. forsythus

and T denticola. However, observation of the cell figures on an individual level

indicates that the numbers of cells specific for P. gingivalis, A.

actinomycetemcomitans and T Denticola were markedly different. With a sample

size of 5, and with large variations between subjects it is difficult to detect statistically

significant differences which could possibly emerge in a study with a larger

population.

Analysis of the results from this study proved interesting. It was felt to be an

achievement to develop a technique such as this one, that has so much potential. It is

a specific detection method, that could potentially provide results using purified

antigens and even specific antigenic epitopes on serial sections of tissue. Molecular

biological work investigating the antigenic target in individuals that have induced a

humoral immune response in periodontal disease could potentially utilise this

technique to test consecutive epitopes of a particular antigen. In addition, progress

with the technique could lead to a potential diagnostic tool. Investigations utilising

this technique to examine the percentage of the plasma cell infiltrate specific for a

particular bacteria, and on a larger scale the percentage that is unidentified is exciting.

The use of purified antigens which are smaller and possibly more sensitive could be

the future for this technique.

270

Chapter 7. Methodological Considerations and General

Conclusions

This chapter attempts to do two things, firstly, it deals with the methodological

considerations and secondly it brings together the results and discussions from the

previous four chapters to a final conclusion.

7.1 Methodological considerations

Sample size

The first study carried out to investigate the effect of periodontal treatment on

antibody titres against a panel of microbial targets was based on a sample size of 25.

This sample size was similar, and in some cases larger, than those of studies

previously published (Tolo et al., 1982; Sandholm and Tolo, 1986; Mouton et al.,

1987; Horibe et al., 1995; Mooney et al., 1995).

The second part of that study investigating the cross-reactivity between micro-

organisms was based on a sample size of 9 people. In addition, the ELISPOT

technique and immunohistochemical studies were carried out both on 5 subjects. The

numbers of subjects involved in these studies are small. It was felt that these numbers

were sufficient to pilot the techniques used, to get a general feel for the results and

observe any trends. However, statistical analysis was difficult with such a small

sample size, and although the most suitable analysis possible was undertaken, the

power of the analysis was weak.

7.2 Conclusions

In chapter 3 the effect of treatment on the serum antibody titres of patients with

chronic periodontitis was examined. The overall conclusion from this study was that

there was no significant difference between the antibody titres measured before and

after periodontal treatment. There could be several reasons for this. Firstly, it could

271

be that the microbial targets used in the study are not causing disease. The panel of

micro-organisms and antigen preparations used were, however, extensive and made

up of a number of the putative periodontal pathogens (Zambon, 1996). Secondly, it

could be that the organisms were not eliminated by the treatment, however, this again

is unlikely as there was significant clinical improvement seen in the patients following

treatment. It could be that the numbers of organisms remained at a level sufficient to

maintain antigenic stimulus but not enough to cause disease, or perhaps that they were

present at sites in the mouth where treatment was not carried out. The reason for not

detecting a change in antibody titre following treatment, could, however, be due to the

time chosen for sampling, hence the length of treatment as discussed in chapter 3. A

final reason which could explain the results obtained is that there could be cross-

reactive antigens that have the ability to maintain levels of antibodies in the absence

of the infecting organisms. The subject of cross-reactivity was investigated therefore,

in the second part of chapter 3.

The second part of chapter 3 examined pre-treatment serum antibody titres in patients

with chronic periodontitis. The study investigated cross-reactivity between antibodies

specific for one periodontal pathogen against three other periodontal micro-organisms.

The results of this experiment suggest that there is cross-reactivity between P.

gingivalis, A. actinomycetemcomitans, P. intermedia and B. forsythus. The results

indicate that when serum is adsorbed against one micro-organism e. g. P. gingivalis,

approximately 60-80% of the antibodies against the other 3 periodontal pathogens (A.

actinomycetemcomitans, P. intermedia and B. forsythus) have also been adsorbed out,

indicating a high level of cross-reactivity between the 4 periodontopathogens. Similar

results were found for all of the pathogens except B. forsythus which seemed to have

less cross-reactivity with the other micro-organisms.

Chapter 3 was an investigation into the systemic humoral immune response of patients

with chronic periodontitis. The systemic immune response is often examined by

observation of serum antibody titres. It is useful as it is thought to reflect the local

immune response and because blood is obtained easily and gives a full patient

immunological assessment as contrasted with GCF which is very much site specific.

However, following the study carried out in chapter 3, the natural progression was to

272

begin to investigate the local immune response, to try and gain a clearer understanding

of the target of the humoral immune response in periodontal disease. Therefore, the

remaining studies described in this thesis concerned the local immune response,

examined by observation of granulation and gingival tissue, removed from

periodontally diseased patients during surgery. Histopathological techniques have

indicated that diseased tissues taken from periodontitis patients are populated by a

large number of B lymphocytes and plasma cells, thus suggesting local antibody

production (Ranney, 1991; Sims et al., 1991). Therefore, the rationale behind the

studies carried out in chapters 4,5 and 6 was that examination of the antibody bearing

cells in the diseased tissues of periodontitis patients will lead to knowledge of the

pathogenic target inducing their production, more efficiently than observation of

antibodies found in the serum. Theories have been put forward (Kinane et al., 1999a)

suggesting possible ways of linking the local and systemic immune responses,

however, they all stem from the induction of an immune response against an antigenic

stimulus in the periodontal tissues.

The general conclusions that were made following the study entitled Ìmmortalised B

cells from periodontal patients' described in chapter 4 are really more a discussion of

the technical complexities of this endeavour and the hypothesis driving this study.

The aim of the study was to produce monoclonal antibodies from the peripheral blood

and granulation tissue of patients with chronic periodontal disease. Expansion of

numbers of peripheral blood B lymphocytes was successful using the CD40 system,

and with both expanded peripheral blood lymphocytes and antibody-bearing cells

isolated from granulation tissue, fusion with a heteromyeloma partner was also

successful. Following fusion, antibody production was monitored and screened for 3-

4 weeks but in all of the 11 experiments carried out, antibody production stopped at 3-

4 weeks and did not return. This is a problem that is often experienced when utilising

this technique as discussed in chapter 4.

The ELISPOT technique was utilised in chapter 5 to investigate the specificity of

antibody bearing cells isolated from the granulation tissue of patients with chronic

periodontal disease. The number of antibody bearing plasma cells specific for the

putative periodontal pathogens P. gingivalis, A. actinomycetemcomitans, P.

273

intermedia and B. forsythus were enumerated. In addition, the number of cells that

were detected against the control, tetanus toxoid, were also counted, to examine the

specificity of the technique. Two different concentrations of cells were used in the

study.

The results showed that the only statistically significant differences in cell number

were between, A. actinomyceteincomitans and tetanus toxoid and B. forsythus and

tetanus toxoid for both cell concentrations. Median values indicated that the number

of cells detected against tetanus toxoid appeared to be much lower than detected

against the periodontal pathogens, however, the differences were not all found to be

statistically significant, probably due to the low sample size of 5.

This study indicated that, the ELISPOT is a technique that can be utilised to study the

target of cells isolated from periodontal tissues and hence, the local immune response

induced in individuals with periodontal disease. In addition, it has suggested that

there is no significant difference between the numbers of cells bearing antibodies to

the four periodontal pathogens P. gingivalis, A. actinomycetemcomitans, P.

intermedia and B. forsythus. Chapter 3 indicated a high percentage of cross-reactivity

between these pathogens. Therefore, we cannot confirm that when an antibody

bearing cell is detected against one micro-organism, it has not been initially induced

against another micro-organism.

For this reason the technique in chapter 6 was developed. The aim of this study was

again the detection of antibody-bearing cells in the tissues of patients with periodontal

disease, however, the technique developed held a number of advantages over that of

the ELISPOT. Firstly, the specially designed 96 well ELISPOT plates were extremely

variable with regard to the bacterial preparations and antigens that were able to bind to

them. Fixed bacteria and the purified antigens tested did not bind successfully to the

plates and the experiments were unsuccessful when these preparations were used. In

addition, enumeration of the ELISPOTs was quite difficult. With regard to the

problem discussed above, the further advantage of the immunohistochemical

technique developed in chapter 6, was that when serial sections were used, it was

274

possible to detect whether antibodies on the surface of plasma cells were binding and

hence cross-reacting with a number of micro-organisms.

The study described in chapter 6 examined paraffin embedded gingival and

granulation tissue sections taken from chronic periodontitis patients. The results

indicated that the number of cells bearing antibody to E. coli was significantly lower

than to the majority of the putative periodontal micro-organisms and antigens tested,

indicating that this technique was specific.

The results of this study firstly indicated that a novel technique had been developed

that was able to detect antibody-bearing cells against whole fixed

periodontopathogens and purified antigens from some of these. The technique was

based on the biotin-avidin system making it inexpensive, easy to use and not labour

intensive. The results indicated that there was no significant difference in the number

of cells detected against any of the whole fixed micro-organisms tested and the

purified antigens leukotoxin (of A. actinomycetemcomitans) and HSP75 (of P.

gingivalis) in either of the tissue types. No cells were detected to be specific against

the ß segment of the RgpA protease (of P. gingivalis).

The results further indicated that there were significantly more cells detected against

P. gingivalis, A. actinomycetemcomitans and P. intermedia in the granulation tissue

compared with the gingival tissue. However, the numbers of cells specific for a

particular antigenic target were then expressed as a percentage of the total plasma cell

infiltrate and when the percentages were observed it became clear that there was no

difference in the percentage of cells specific to a particular micro-organism in the

different tissue types. The difference lay in the size of the infiltrate and, therefore, the

number of plasma cells that had permeated into the tissues. These results were,

therefore, consistent with the ELISPOT results which suggested that in general there

is no significant difference between the number of cells bearing antibody to the

different micro-organisms and antigens tested. The percentages of plasma cells

specific for each of the targets were presented in the form of a pie chart for each

subject individually. It was very interesting that the results could be expressed in this

form, firstly because it gives the reader a clearer idea of the number of cells specific

275

for a particular target, and secondly because it allows the percentage of plasma cells

whose target is unidentified to be highlighted.

7.3 Suggestions for future research

7.3.1 Examination of the effect of treatment on the humoral immune response

As discussed in chapter 3, it is very difficult to compare studies performed by

different groups, as the length of treatment appears to vary so much. To truly measure

the effects of treatment on the immune response, it would be interesting to analyse

serum samples at much shorter time intervals following treatment. For example if a

serum sample was taken every 2-4 weeks following treatment for a time period of 6

months. Examination of samples taken like this would allow the dynamics i. e. any

increases or decreases in titre to be noted over time.

7.3.2 Further work on monoclonal antibody production

The aim of the monoclonal antibody production study was novel but proved too

complex at this time. Obtaining monoclonal antibodies produced by the protocol

described in chapter 4 could in future provide some interesting results. If monoclonal

antibodies were produced and the antigen binding portion of these sequenced, it could

lead to identification of the antigenic target that has induced the immune response,

whether it be a known antigen or not.

7.3.3 Detection of antibody-bearing plasma cells in granulation and gingival tissue

The technique developed in this study has potential and should be utilised in studies of

other chronic infectious diseases not only periodontitis. In terms of the study

described in chapter 6, it would be interesting to compare tissue taken from patients

with different disease forms (e. g. gingivitis, aggressive periodontitis, chronic

periodontitis), and see if there are differences in the immune infiltrate between these

different disease entities. In addition, analysis of immunoglobulin isotypes and

examination of further antigens would be of interest. Cross-reactivity studies could

276

also be carried out by utilising serial sections. This technique could also be used to

further observe the percentages of the plasma cell infiltrate specific to a particular

antigenic target. Having the ability to be able to do this could prove valuable in the

future as it puts the results into a more understandable context. The technique should

be utilised to look at a larger panel of purified antigens. A study specifically

investigating the phenomena of cross-reactivity is also very important and this

technique would be an excellent tool to do this with.

7.4 Concluding remarks

The future identification of the target or targets of the humoral immune response in

periodontal disease could have wide-ranging implications. They could potentially

include the possibility of new diagnostics, prophylactics and therapeutics. Whether

antibody responses to specific antigens are markers of disease susceptibility or activity

is currently unclear. Progress is, however, being made in identifying marker antigens

and species out of the 300-400 bacterial species identified in microbial plaque.

The studies presented in this thesis have provided an insight into how the antigenic

targets of plasma cells in periodontitis may be elucidated. The results reported show

the development of a new detection technique that can be used to identify plasma cells

directly that are bearing antibodies to a particular antigenic target. The results also

emphasise the uses of techniques such as the ELISPOT, another method that can be

utilised to investigate the target of the immune response in periodontal disease. As

well as highlighting aspects regarding technical issues, the studies reported in this

thesis have addressed the issues of cross-reactivity between Gram negative bacteria,

the target of antibodies being produced locally in diseased tissues of periodontitis

patients and responses to antigens that have previously been described as

immunodominant such as LPS, leukotoxin and HSP.

277

Bibliography

Accolla RS, Adorini L, Sartoris S, Sinigaglia F, Guardiola J. MHC: orchestrating the

immune response. Immunol Today 1995; 16: 8-11.

Aduse-Opoku J, Muir J, Slaney JM, Rangarajan M, Curtis MA. Characterisation,

genetic analysis, and expression of a protease antigen (PrpPI) of Porphyromonas

gingivalis W50. Infect Immun 1995; 63: 4744-4754.

Agarwal BB, Eessalu TE, Hass PE. Characterisation of receptors for human tumor

necrosis factorand their regulation by y-interferon. Nature 1985; 318: 665-667.

Agarwal S, Huang JP, Piesco NP, Suzuki JB, Riccelli AE, Johns LP. Altered

neutrophil function in localized juvenile periodontitis: intrinsic or induced? J

Periodontol 1996; 67: 337-344.

Ah MK, Johnson GK, Kaldahl WB, Patil KD, Kalkwarf KL. The effect of smoking on

the response to periodontal therapy. J Clin Periodontol 1994; 21: 91-97.

Ainamo J, Löe H. Anatomical characterisation of gingiva. A clinical and microscopic

study of the free and attached gingiva. J Periodontol 1966; 37: 5-13.

Alaluusua S, Asikainen S. Detection and distribution of Actinobacillus

actinomycetemcomitans in the primary dentition. J Periodontol 1988; 59: 504-507.

Albeda S, Smith C, Ward P. Adhesion molecules and inflammatory injury. FASEB J

1994; 8: 504-512.

Aldred MJ, Bartold PM. Genetic disorders of the gingivae and periodontium.

Periodontol 2000 1998; 18: 7-20.

278

Aleo JJ, Denranzio FA, Farber PA, Varboncoeur AP. The presence and biological

activity of cementum endotoxin. J Periodontol 1974; 45: 672-675.

Aleo JJ. Stimulation of macromolecular synthesis by endotoxin treated 3T6

fibroblasts. Experientia 1980; 36: 546-547.

Alexander MB, Damoulis PD. The role of cytokines in the pathogenesis of

periodontal disease. Curr Opin Periodontol 1994; 39-53.

Allan RB, Wilkinson PC. 1978. A visual analysis of chemotactic and chemokinetic

locomotion of human neutrophil leucocytes. Exp Cell Res 1978; 111: 191-203.

Allen PM. Antigen processing at the molecular level. Immunol Today 1987; 8: 270-

273.

Altman LC, Page RC, Ebersole JL, Vandesteen EG. Assessment of host defences and

serum antibodies to suspected periodontal pathogens in patients with various types of

periodontitis. J Periodontal Res 1982; 17: 495-497.

Aman P, Ehlin-Henriksson B, Klein G. Epstein-Barr virus susceptibility of normal

human B lymphocyte populations. J Exp Med 1984; 159: 208-220.

Andersson J, Melchers F. The antibody repertoire of hybrid cells obtained by fusion of

X63-AG8 myeloma cells with mitogen-activated B cell blasts. Curr Top Micobiol

Immunol 1978; 81: 130-138.

Ando T, Kato T, Ishihara K, Ogiuchi H, Okuda K. Heat Shock Proteins in the Human

Periodontal Disease Process. Microbiol Immunol 1995; 39: 321-327.

Anerud A, Löe H, Boysen H. Smith M. The natural history of periodontal disease in

man. Changes in gingival health and oral hygiene before 40 years of age. J Periodontal

Res 1979; 14: 526-540.

279

Angelopoulaas AP. Studies of mast cells in the human gingiva. J Periodontal Res

1973; 8: 314-322.

Anusaksathien 0, Dolby AE. Autoimmunity in periodontal disease. J Oral Pathol Med

1991; 20: 101-107.

Aoyagi T, Sugawara-Aoyagi M, Yamazaki K, Hara K. Interleukin-4 (IL-4) and IL-6-

producing memory T cells in peripheral blood and gingival tissues in periodontitis

patients with high serum antibody titres to Porphyromonas gingivalis. Oral Microbiol

Immunol 1995; 10: 304-310.

Apel M, Berek C. In: Berrih-Aknin ES, Imhof B, editors. Lymphatic tissues and in

vivo immune responses. New York: Dekker, 1991: 355-359.

Arafat AH. Periodontal status during pregnancy. J Periodontol 1974; 45: 641-643.

Arno A, Schei 0, Lftvdal A, Waerhaug J. Alveolar bone loss as a function of tobacco

consumption. Acta Odontol Scand 1959; 17: 3-10.

Asikainen S, Lai C-H, Alaluusua S, Slots J. Distribution of Actinobacillus

actinomycetemcomitans serotypes in periodontal health and disease. Oral Microbiol

Immunol 1991; 6: 115-118.

Astemborski JA, Boughman JA, Myrick PO, et al. Clinical and laboratory

characterization of early onset periodontitis. J Periodontol 1989; 60: 557-563.

Atlaw T, Kozbor D, Roder JC. Human monoclonal antibodies against Mycobacterium

leprae. Infect Immun 1985; 49: 104-110.

Aukhil I, Lopatin DE, Syed SA, Morrison EC, Kowalski CJ. The effects of

periodontal therapy on serum antibody (IgG) levels to plaque micro-organisms. J Clin

Periodontol 1988; 15: 544-550.

280

Babbitt B, Allen P, Matsueda G, Haber E, Unanue E. Binding of immunogenic

peptides to la histocompatability molecules. Nature 1985; 317: 359-361.

Bacic M, Plancak D, Granic M. CPITN assessment of periodontal status in diabetics. J

Periodontol 1988; 59: 816-822.

Badersten A, Nilveus R, Egelberg J. Effect of non-surgical periodontal therapy. I.

Moderately advanced periodontitis. J Clin Periodontol 1981; 8: 57-72.

Baehni P, Tsai CC, McArthur WP, Hammond BF, Taichman NS. Interaction of

inflammatory cells and oral microorganisms. VIII. Detection of leukotoxic activity of

a plaque-derived gram-negative microorganisms. Infect Immun 1979; 24: 233-243.

Baehni P, Tsai CC, McArthur WP, Hammond BF, Socransky SS, Taichman NS.

Leukotoxicity of Various Actinobacilli actinomycetemcomitans isolates. J Dent Res

1980; 59A: 323. Abstract.

Baehni PC, Guggenheim B. Potential of diagnostic microbiology for treatment and

prognosis of dental caries and periodontal diseases. Crit Rev Oral Biol Med 1996; 7:

237-258.

Banchereau J, Paoli PP. Valle A, Garcia E, Rousset F. Long-term human B cell lines

dependent on interleukin-4 and antibody to CD40. Science 1991; 251: 70-72.

Banchereau J, Rosset F. Growing human B lymphocytes in the CD40 system. Nature

1991; 353: 678-679.

Banchereau J, Bazan F, Blanchard D, et al. The CD40 antigen and its ligand. Annu

Rev Immunol 1994: 12: 881-922.

Bangsbourg JM, Collins MT, Holby N, Hindersson P. Cloning and expression of the

Legionella micdadei "common antigen" in Escherichia coli. APMIS 1989; 97: 14-22.

281

Baranowska HI, Palmer RM, Wilson RF. A comparison of antibody levels to

Bacteroides gingivalis in serum and crevicular fluid from patients with untreated

periodontitis. Oral Microbiol Immunol 1989; 4: 173-175.

Barthold PM, Haynes DR. Interleukin-6 production by human gingival fibroblasts. J

Periodontal Res 1991; 26: 339-345.

Bartolucci EG, Parkes RB. Accelerated periodontal break down in uncontrolled

diabetics. Pathogenesis and treatment. Oral Surg Oral Med Oral Pathol Oral Radio]

Endod 1981; 52: 387-390.

Beachey EH. Bacterial adherance: adhesion-receptor interactions mediating the

attachment of bacteria to mucosal surfaces. J Infect Dis 1981; 143: 325-345.

Beck JD, Koch GG, Rozier RG, Tudor GE. Prevalence and risk indicators for

periodontal attachment loss in a population of older community-dwelling blacks and

whites. J Periodontol 1990; 61: 521-528.

Beck J, Garcia R, Heiss G, Vokonas PS, Offenbacher S. Periodontal disease and

cardiovascular disease. J Periodontol 1996; 67: 1123-1137.

Becks H. Normal and pathologic pocket formation. J Am Dent Assoc 1927; 16: 2167-

2188.

Bedi GS. Purification and characterization of lysine- and arginine-specific gingivain

proteases from Porphyromonas gingivalis. Prep Biochem 1994; 24: 251-261.

Belting CM, Hiniker JJ, Dummett CO. Influence of diabetes mellitus on the severity

of periodontal disease. J Periodontol 1964; 35: 476-480.

Bendtzen K. Cytokines and natural regulators of cytokines. Immunol Lett 1994; 43:

111-123.

282

Benjamini E, Sunshine G, Leskowitz S. Immunology. A short course. Third edition.

New York: Wiley-Liss, 1996.

Berek C. The development of B cells and B-cell repertoire in the microenvironment of

the germinal center. Immunol Rev 1992; 126: 5-19.

Berek C, Ziegner M. The maturation of the immune response. Immunol Today 1993;

14: 400-404.

Bergstrom J, Floderus-Myehed B. Co-twin control study of the relationship between

smoking and some periodontal disease factors. Community Dent Oral Epidemiol

1983; 11: 113-116.

Bergström J. Cigarette smoking as a risk indicator in chronic periodontal disease.

Community Dent Oral Epidemiol 1989; 17: 245-247.

Berglund Se. Immunoglobulin in human gingiva with specificity for oral bacteria. J

Periodontol 1971; 42: 546-551.

Berkowitz R, Jordan V. Similarity of bacteriocins of Streptococcus mutans from

mother and infant. Arch Oral Biol 1975; 20: 725-730.

Beutler B, Mahoney J, Le Trang N, Pekala P, Cerami A. Purification of cachectin, a

lipoprotein lipase-suppressing hormone secreted by endotoxin-induced RAW 264.7

cells. J Exp Med 1985; 161: 984-995.

Bevilaqua MP. Circulation 1989; 80: ii-i.

Bevilaqua MP, Butcher E, Furie B, et al. Selectins: A family of adhesion receptors.

Letters to the Editor. Cell 1991; 67: 233.

283

Bhat M. Periodontal health of 14-17 year old US school children. J Public Health

Dent 1991; 51: 5-11.

Birkedal-Hansen H. Roles of cytokines and inflammatory mediators in tissue

destruction. J Periodontal Res 1993; 28: 500-510.

Bissada NF, Manouchehr-Pour M, Haddow M, Spagnuolo PJ. Neutrophil functional

activity in juvenile and adult onset diabetic patients with mild and severe

periodontitis. J Periodontal Res 1982; 17: 500-502.

Block PJ, Yoran C, Fox AC, Kaltman AJ. Actinobacillus actinonrycetemcomitans

endocarditis: Report of a case and review of the literature. Am J Med Sci 1973; 266:

387-392.

Bloom BR, Modlin RJ, Salgame P. Stigma variations: observations on suppressor T

cells and leprosy. Annu Rev Immunol 1992; 10: 453-488.

Blum JS, Ma C, Kovats S. Antigen-presenting cells and the selection of

immunodominant epitopes. Crit Rev Immunol 1997; 17: 411-417.

Boehringer H, Berthold PH, Taichman NS. Studies on the interaction of human

neutrophils with plaque spirochetes. J Periodontal Res 1986; 21: 195-209.

Bogard WC, Hornberger E, Kung PC. Production and characterisation of human

monoclonal antibodies against Gram negative bacteria. In: Engleman EG, Foung

SKH, Larrick J, Raubitschek A, editors. Human hybridomas and monoclonal

antibodies. New York: Plenum, 1985: 95.

Bolstad Al, Kristoffersen T, Olsen I, et al. Outer membrane proteins of Actinohacillus

actinomycetemcomitans and Haemophilus aphrophilus studied by SDS-PAGE and

immunoblotting. Oral Microbiol Immunol 1990; 5: 155-161.

284

Booth V, Lehner T. Characterisation of the Porphyromonas gingivalis antigen

recognized by a monoclonal antibody which prevents colonization by the organism. J


Borrebaeck CAK, Danielsson L, Möller SA. Human monoclonal antibodies produced

by primary in vitro immunization of peripheral blood lymphocytes. Proc Natl Acad

Sci USA 1988; 85: 3995-3999.

Borrebaeck CAK. Strategy for the production of human monoclonal antibodies using

in vitro activated B cells. J Immunol Methods 1989; 123: 157-165.

Boughman JA, Halloran SL, Roulsten D et al. An autosomal dominant form of

juvenile periodontitis: its localization to chromosome 4 and linkage to dentinogenesis

imperfecta and Gc. J Craniofac Genet Dev Biol 1986; 6: 341-350.

Bowers GM. A study of the width of attached gingiva. J Periodontol 1963; 34: 201-

209.

Box HK. Studies in periodontal pathology. Toronto: Canadian Dental Research

Foundation, 1924.

Boyd JE, Hastings I, Farzad Z, James K, McClelland DBL. Experiences in the

production of human monoclonal antibodies to tetanus toxoid. Dev Biol Stand 1984;

57: 93-98.

Brandtzaeg P. Immunochemical comparison of proteins in human gingival pocket

fluid, serum and saliva. Arch Oral Biol 1965; 10: 795-803.

Brandtzaeg P, Kraus FW. Autoimmunity and periodontal disease. Odontologisk

Tidskrift 1965; 73: 281-293.

285

Brandtzaeg P. Local factors of resistance in the gingival area. J Periodontal Res 1966;

1: 19-42.

Brandtzaeg P. Local formation and transport of immunoglobulins related to the oral

cavity. In: Macphee T, editor. Host resistance to commensal bacteria. Edinburgh:

Churchill Livingstone, 1972: 116-150.

Brandtzaeg P. Immunology of inflammatory periodontal lesions. Int Dent J 1973; 23:

438-454.

Brandtzaeg P, Tolo K. Immunoglobulin systems of the gingiva. In: Lehner T, editor.

The borderland between caries and periodontal disease. London: Academic Press,

1977: 145-184.

Brant EE, Sojar HT, Sharma A, Bedi GS, Genco RJ, DeNardin E. Identification of

linear antigenic sites of Porphyromonas gingivalis 43kDa fimbrillin subunit. Oral

Microbiol Immunol 1995; 10: 146-150.

Brecx MC, Gautschi M, Gehr P, Lang NP. Variability of histologic criteria in

clinically healthy human gingiva. J Periodontal Res 1987; 22: 468-472.

Brecx M, Frölicher I, Gehr P, Lang NP. Stereological observations on long term

experimental gingivitis in man. J Clin Periodontol 1988; 15: 621-627.

Bretscher PA. The regulatory functions of CD4+ and CD8+ T-cell sunsets in immune

regulation. Res Immunol 1991; 142: 45-50.

Brill N, Brönnestam R. Immuno-electophoretic study of tissue fluid from gingival

pockets. Acta Odontol Scand 1960; 18: 95-100.

286

Briskin M, Winsor-Hines D, Shyjan A, et al. Human Mucosal Addressin Cell

Adhesion Molecule-I is preferentially expressed in intestinal tract and assocaited

lymphoid tissue. Am J Pathol 1997; 151: 97-110.

Bron D, Feinberg MB, Teng NNH, Kaplan HS. Production of human monoclonal IgG

antibodies against Rhesus (D) antigen. Proc Natl Acad Sci USA 1984; 81: 3214-3217.

Buchmeier N, Heffron F. Induction of Salmonella stress proteins upon infection of

macrophages. Science 1990; 248: 730-732.

Buerki H, Cottier H, Hess MW, Laissue J, Stoner RD. Distinctive medullary and

germinal center proliferative patterns in mouse lymph nodes after regional primary

and secondary stimulation with tetanus toxoid. J Immunol 1974; 112: 1961-1970.

Burnett KG, Leung JP, Martinis J. Human monoclonal antibodies to defined antigens:

Towards clinical applications. In: Engleman EG, Foung SKH, Larrick J, Raubitschek

A Editors. Human hybridomas and monoclonal antibodies. New York: Plenum, 1985

113.

Butcher EC, Rouse RV, Coffman RL, Nottenburg CN, Hardy R, Weissman IL.

Surface phenotype of peyer's patch germinal center cells: implications for the role of

germinal centers in B cell differentiation. J Immunol 1982; 129: 2698-2707.

Buus S, Sette A, Colon S, Miles C, Grey H. The relationship between major

histocompatability complex (MHC) restriction and the capacity of la to bind

immunogenic peptides. Science 1987; 235: 1353-1358.

Califano JV, Schenkein HA and Tew JG. Immunodominant antigen of Actinobacillus

actinomycetemcomitans Y4 in high-responder patients. Infect Immun 1989; 57: 1582-

1589.

287

Califano JV, Schenkein HA, Tew JG. Immunodominant antigens of A.

actinomycetemcomitans serotype b in early onset periodontitis patients. Oral


Califano JV, Pace BE, Gunsolley JC, Shenkein HA, Lally ET, Tew JG. Antibody

reactive with Actinobacillus actinomycetemcomitans leukotoxin in early-onset

periodontitis patients. Oral Microbiol Immunol 1997; 12: 20-26.

Campbell M. Epidemiology of periodontal disease in the diabetic and non-diabetic.

Aust Dent J 1972; 17: 274-278.

Campling BG, Cole SPC, Atlaw T et al. Practical aspects of human-human

hybridomas. In: Schook LB editor. Monoclonal antibody production techniques and

applications. New York: Marcel Dekker Inc., 1987: 4-23.

Carvel R, Can RF. Clinical-pathologic correlations of 385 lesions of marginal

destructive periodontitis. J Periodontol 1982; 53: 328-333.

Castelli WA. Vascular architecture of the human adult mandible. J Dent Res 1963; 42:

786-792.

Celenligil H, Kansu E, Eratalay K. Juvenile and rapidly progressive periodontitis.

Peripheral blood lymphocyte populations. J Clin Periodontol 1990; 17: 207-210.

Celenligil H, Kansu E, Ruacan S, Eratalay K, Caglayan G. In situ characterization of

gingival mononuclear cells in rapidly progressive periodontitis. J Periodontol 1993;

64: 120-127.

Celenligil H, Ebersole JL. Characteristics and responses of EBV immortalized B cells

from periodontal disease patients. Oral Dis 1997; 3: 262-271.

288

celenligil H, Ebersole JL. Analysis of serum antibody responses to

periodontopathogens in early-onset periodontitis patients from different geographical

locations. J Clin Periodontol 1998; 25: 994-1002.

Chan MA, Stein LD, Dosch HM, Sigal NH. Heteogeneity of EBV-transformable

human B lymphocyte populations. J Immunol 1986; 136: 106-112.

Chen CC, Chang KL, Huang JF, Huang JS, Tsai CC. Correlation of interleukin-1 beta,

interleukin-6 and periodontitis. Kao Hsiung I Hseu Ko Hseueh Tsa Chih 1997; 13:

609-617.

Chen HA, Johnson BD, Sims TJ, et al. Humoral immune responses to Porphvromonas

gingivalis before and following therapy in rapidly progressive periodontitis patients. J

Periodontol 1991; 62: 781-791.

Chen HA, Weinberg A, Darveau RP, Engel D, Page RC. Immunodominant antigens

of Porphyromonas gingivalis in Patients with Rapidly Progressive Periodontitis. Oral


Chen Y, Boros DL. Identification of the immunodominant T cell epitope of p38, a

major egg antigen, and characterization of the epitope-specific Th responsiveness

during murine Shistosomiasis mansoni. J Immunol 1998; 160: 5420-5427.

Chen Z, Potempa A, Wilkstrom M, Travis J. Purification and characterization of a 50-

kDa cysteine proteinase (gingipain) from Porphyromonas gingivalis. J Biol Chem

1992; 267: 18896-18901.

Choi JI, Nakagawa T, Yamada S, Takazoe I, Okuda K. Clinical, microbiological and

immunological studies on recurrent periodontal disease. J Clin Periodontol 1990; 17:

426-434.

289

Christersson LA, Slots J, Rosling BG, Genco RJ. Microbiological and clinical effects

of surgical treatment of localized juvenile periodontitis. J Clin Periodontol 1985; 12:

465-476.

Chu L, Ebersole JL, Holt SC. Cystalysin, a 46-kilodalton cysteine desulfhydrase from

Treponema denticola, with hemolytic and hemoxidative activities. Infect Immun

1997; 65: 3231-3238.

Cianciola U, Park BH, Bruck E, Mosovich L, Genco R. Prevalence of periodontal

disease in insulin-dependent diabetes mellitus (juvenile diabetes). J Am Dent Assoc

1982; 104: 653-660.

Cimasoni G. Crevicular fluid updated. Monogr Oral Sci 1983; 12: 1-152.

Cisar JO, Vatter AE. Surface fibrils (fimbriae) of Actinoniyces viscosus TI4V. Infect

Immun 1979; 24: 523-531.

Clark WB. Actinomyces fimbriae and adherance to hydroxyapatite. In: Mergenhagen

SE, Rosan B, editors. Molecular basis of oral microbial adhesion. Washington DC:

American Society for microbiology, 1985: 103-108.

Cobb CM, Singla 0, Feil PH, Theisen FC, Schultz P. Comparison of NK-cell (leu-7+

and Leu-1lb+) populations in clinically healthy gingivae, chronic gingivitis and

chronic adult periodontitis. J Periodontal Res 1989; 24: 1-7.

Cohen B. Morphological factors in the pathogenesis of periodontal disease. Br Dent j

1959; 107: 31-39.

Condorelli F, Scalia G, Cali G, Rossetti B, Nicoletti G, Lo Bue AM. Isolation of

Porphyromonas gingivalis and detection of Immunoglobulin A specific to fimbrial

antigen gingival crevicular fluid. J Clin Microbiol 1998; 36: 2322-2325.

290

Costerton JW, Lewandowski Z, Caldwell DE, Korber DR, Lappin-Scott HM.

Microbial biofilms. Annu Rev Microbiol 1995; 49: 711-745.

Cote RJ, Morrissey DM, Houghton AN, Beattie Jr. EJ, Oettgen HF, Old U.

Generation of human monoclonal antibodies reactive with cellular antigens. Proc Natl

Acad Sci USA 1983; 80: 2026-2030.

Cote RJ, Morrissey DM, Oettgen HF, Old U. Analysis of human monoclonal

antibodies derived from lymphocytes of patients with cancer. Fed Proc 1984; 43:

2465-2469.

Cote RJ, Morrissey DM, Houghton AN, et al. Specificity analysis of human

monoclonal antibodies reactive with cell surface and intracellular antigens. Proc Natl

Acad Sci USA 1986; 83: 2959-2963.

Cotran RS, Gimbrone Jr MA, Bevilaqua MP, Mendrick DL, Pober JS. Induction and

detection of a human endothelial antigen in vivo. J Exp Med 1986; 164: 661-666.

Cox RJ, Brokstad KA, Zuckerman MA, Wood JM, Haaheim LR, Oxford JS. An early

humoral immune response in peripheral blood following parenteral inactivated

influenza vaccination. Vaccine 1994; 12: 993-999.

Crawford JM, Watanabe K. Cell adhesion molecules in inflammation and immunity:

Relevance to periodontal diseases. Crit Rev Oral Biol Med 1994; 5: 91-123.

Cridland JC, Booth V, Ashley FP, Curtis MA, Wilson RF, Shepherd P. Preliminary

characterisation of antigens recognised by monoclonal antibodies raised to

Porphyromonas gingivalis and by sera from patients with periodontitis. J Periodontal

Res 1994; 29: 399-347.

Croce CM, Linnenbach A, Hall W, Steplwski Z, Koprowski II. Production of human

hybridomas secreting antibodies to measles virus. Nature 1980; 288: 488-489.

291

Cui Y, Chang L-J. Computer assisted, quantitatiove cytokine enzyme-linked immunospot analysis of human immune effector cell function. Biotechniques 1997;

22: 1146-1149.

Cullinan MO, Sachs J, Wolf E, Seymour GJ. The distribution of HLA-A and -B

antigens in patients and their families with periodontitis. J Periodontal Res 1980; 15:

177-184.

Cunningham MD, Seachord C, Ratcliffe K, Bainbridge B, Aruffo A, Darveau RP.

Helicobacter pylori and Porphyromonas gingivalis lipopolysaccharides are poorly

transferred to recombinant soluble CD14. Infect Immun 1996; 64: 3601-3608.

Cupps TR, Goldsmith PK, Volkman DJ, Gerin JL, Purcell RH, Fauci AS. Activation

of human peripheral blood B cells following immunization with hepatitis B surface

antigen vaccine. Cell Immunol 1984; 86: 145-154.

Curtis MA, Ramakrishnan M, Slaney JM. Characterisation of the Trypsin-like

Enzymes of Porphyromonas gingivalis W83 using a Radio-labelled Active-site-

directed Inhibitor. J Gen Microbiol 1993; 139: 949-955.

Curtis MA, Aduse-Opoku J, Slaney JM, et al. Characterisation of an Adherence and

Antigenic Determinant of the Arg-1 Protease of Porphyromonas gingivalis which is

present on Multiple Gene Products. Infect Immun 1996; 64: 2532-2539.

Cutler CW, Kalmar JR, Arnold RR. Phagocytosis of virulent Porphyromonas

gingivalis by human polymorphonuclear leukocytes requires specific immunoglobulin

G. Infect Immun 1991 a; 59: 2097-2104.

Cutler CW, Kalmar JR, Arnold RR. Antibody-dependent alternate pathway of

complement activation in opsonophagocytosis of Porphyromonas gingivalis. Infect

Immun 1991b; 59: 2105-2109.

292

Cutler CS, Eke P, Arnold RR, Van Dyke TE. Defective neutrophil function in an

insulin-dependent diabetes mellitus patients. A case report. J Periodontol 1991c; 62:

394-401.

Cutler CW, Arnold RR, Schenkein HA. Inhibition of C3 and IgG Proteolysis enhances

phagocytosis of Porphyromonas gingivalis. J Immunol 1993; 151: 7016-7029.

Czerkinsky CC, Nilsson LA, Nygren H, Ouchterlony 0, Tarkowski A. A solid-phase-

enzyme-linked immunospot (ELISPOT) assay for enumeration of specific antibody-

secreting cells. J Immunol Methods: 1983; 65: 109-121.

Dahlen G. Role of suspected periodontopathogens in microbiological monitoring of

periodontitis. Adv Dent Res 1993; 7: 163-174

Damle NK, Doyle LV. Ability of human T lymphocytes to adhere to vascular

endothelial cells and to augment endothelial permeability to macromolecules is linked

to their state of posthymic maturation. J Immunol 1990; 144: 1233-1240.

Daniel MA, Van Dyke TE. Alterations in phagocyte function and periodontal

infection. J Periodontol 1996; 67: 1070-1075.

Daniel WW. Applied nonparametric statistics. London: Houghton Mifflin, 1978: 231-

397.

Darveau RP, Hancock REW. Procedure for isolation of lipopolysaccharides from both

rough and smooth Pseudomonas aeruginosa and Salmonella typhimurium strains. J

Bacteriol 1983; 155: 831-838.

Darveau RP, Cunningham MD, Bailey T, et al. Ability of bacteria associated with

chronic inflammatory disease to stimulate E-selectin expression and promote

neutrophil adhesion. Infect Immun 1995; 63: 1311-1317.

293

Darveau RP, Tanner A, Page RC. The microbial challenge in periodontitis.

Periodontol 2000 1997; 14: 12-32.

Davidson RL, Gerald PS. Improved techniques for the induction of mammalian cell

hybridisation by polyethylene glycol. Somat Cell Mot Genet 1976; 2: 165-170.

DeFougerolles AR, Stacker SA, Schwarting R, Springer TA. Characterization of

ICAM-2 and evidence for a third counter receptor for LFA-1. J Exp Med 1991; 174:

253-267.

DeFougerolles AR, Springer TA. Intercellular adhesion molecule 3, a third adhesion

counter-receptor for lymphocyte function-associated molecule-I on resting

lymphocytes. J Exp Med 1992; 175: 185-190.

Delacroix DL, Dive C, Rambaud C, Vaerman JP. IgA subclasses in various secretions

and in serum. Immunology 1982; 47: 383-385.

De Panfilis G, Manara GC, Ferrari C, Torresani C. Adhesion molecules on the plasma

membrane of epidermal cells. IT. The intercellular adhesion molecule-I is

constitutively present on the cell surface of human resting langerhans cells. J Invest

Dermatol 1990; 94: 317-321.

Dewhirst FE, Paster BJ. DNA probes analyses for the detection of periodontopathic

bacteria in clinical samples. In: Hamada S, Holt SC, McGhee J, editors. Periodontal

disease. Pathogens and host immune responses. Tokyo: Quintessence publishing Co.,

Ltd., 1991: 367-378.

Diamond MS, Staunton DE, De Fougerolles AR, et al. ICAM-1 (CD54)- as counter-

receptor for Mac-1 (CD1 lb/CD18). J Cell Biol 1990; 111: 3129-3139.

Dibart S, Skobe Z, Snapp KR, Socransky SS, Smith CM, Kent Jr RL. Identification of bacterial species on or in crevicular epithelial cells from healthy and periodontally

294

diseased patients using DNA-DNA hybridisation. Oral Microbiol Immunol 1998; 13:

30-35.

DiFranco CF, Toto PD, Rowden G, Gargiulo AW, Keene JJ Jr., Connely E.

Identification of langerhans cells in human gingival epithelium. J Periodontol 1985;

56: 48-54

DiRenzo JM, McKay TL. Identification and characterisation of genetic cluster groups

of Actinobacillus actinomycetemcomitans isolated from the human oral cavitiy. J Clin

Microbiol 1994; 32: 75-81.

Doty SL, Lopatin DE, Syed SA, Smith FN. Humoral immune response to oral

microorganisms in periodontitis. Infect Immun 1982; 37: 499-505.

Dreizen S, McCredie KB, Keating MJ. Chemotherapy associated oral hemorrhages in

adults with acute leukaemia. Oral Surg Oral Med Oral Pathol 1984; 57: 494-498.

Duan Y, Fisher E, Malamud D, Golub E, Demuth DR. Calcium-binding properties of

SSP-5, the Streptococcus gordonii M5 receptor for salivary agglutinin. Infect Immun

1994; 62: 5220-5226.

Dustin ML, Rothlein R, Bhan AK, Dinarello CA, Springer TA. Induction by IL-1 and

interferon-y tissue distribution, biochemistry, and function of a natural adherence

molecule (ICAM-1). J Immunol 1986; 137: 245-254.

Dzink JL, Socransky SS, Ebersole JL, Frey DE. ELISA and conventional techniques

for identification of black-pigmented Bacteroides isolated from periodontal pockets. J


Dzink JL, Socransky SS, Haffajee AD. The predominant cultivable microbiota of

active and inactive lesions of destructive periodontal diseases. J Clin Periodontol

1988; 15: 316-323.

295

Dzink JL, Gibbons RJ, Childs III WC, Socransky SS. The predominant cultivable

microbiota of crevicular epithelial cells. Oral Microbiol Immunol 1989; 4: 1-5.

Ebersole JL, Frey DE, Taubman MA, Smith DJ. An ELISA for measuring serum

antibodies to Actinobacillus actinomycetemcomitans. J Periodontal Res 1980; 15:

621-632.

Ebersole JL, Taubman MA, Smith DJ, Socransky SS. Humoral immune responses and diagnosis of human periodontal disease. J Periodontal Res 1982a; 17: 478-480.

Ebersole JL, Taubman MA, Smith DJ, Genco RJ, Frey DE. Human immune responses

to oral microorgansims. I. Association of localized juvenile periodontitis (LJP) with

serum antibody responses to Actinobacillus actinomycetemcomitans. Clin Exp

Immunol 1982b; 47: 43-52.

Ebersole JL, Taubman MA, Smith DJ, Hammond BF, Frey DE. Human immune

responses to oral micro-organisms. II serum antibody responses to antigens from

Actinobacillus actinomycetemcomitans and the correlation with localised juvenille

periodontitis. J Clin Immunol 1983; 3: 321-331.

Ebersole JL, Taubman MA, Smith DJ. Local antibody responses in periodontal

diseases. J Periodontol 1985a; 56: 51-55.

Ebersole JL, Taubman MA, Smith DJ. Gingival crevicular fluid antibody to oral

microorganisms II. Distribution and specificity to local antibody reponses. J

Periodontal Res 1985b; 20: 349-356.

Ebersole JL, Frey DE, Taubman MA, Smith DJ. IgG subclass antibody responses to

periodontal microorganisms. J Dent Res 1985c; 64: 197.

296

Ebersole JL, Taubman MA, Smith DJ, Haffajee AD. Effect of subgingival scaling on

systemic antibody responses to oral microorganisms. Infect Immun 1985d; 48: 534-

539.

Ebersole JL, Taubman MA, Smith DJ, Frey DE, Haffajee AD, Socransky SS. Human

serum antibody responses to oral microorganisms. IV. Correlation with homologous

infection. Oral Microbiol Immunol 1987; 2: 53-59.

Ebersole JL, Holt SC. Serum antibodies to periodontopathic microorganisms: specific

induction. In: Guggenheim B, editor. Periodontology Today. Basel: Karger, 1988:

169.

Ebersole JL, Steffen MJ, Sandoval M-N. Quantification of IgG subclass antibodies to

periodontopathogens. J Dent Res 1989; 68: 222.

Ebersole JL. Systemic humoral immune responses in periodontal disease. Crit Rev

Oral Biol Med 1990; 1: 283-331.

Ebersole JL, Holt SC. Immunological procedures for diagnosis and risk assessment in

periodontal diseases. In: Johnson NW, editor. Risk markers for oral diseases. Volume

3: Periodontal diseases, markers of susceptibility and activity. Cambridge: Cambridge

University Press, 1991: 203-227.

Ebersole JL, Sandoval M-N, Steffen MJ, Cappelli D. Serum antibody to

Actinobacillus actinomycetemcomitans-infected patients with periodontal disease.

Infect Immun 1991; 59: 1795-1802.

Ebersole JL, Cappelli D, Steffen MJ. Characteristics and utilization of antibody

measurements in clinical studies of periodontal disease. J Periodontol 1992; 63: 1110-

1116.

297

Ebersole JL, Cappelli D. Gingival crevicular fluid antibody to A.

actinomycetemcomitans in periodontal disease. Oral Microbiol Immunol 1994; 9: 335-

344.

Ebersole JL. Steffen MJ. Human antibody responses to outer envelope antigens of

Porphyromonas gingivalis serotypes. J Periodontal Res 1994; 30: 1-14.

Ebersole JL, Taubman MA. The protective nature of host responses in periodontal

diseases. Periodontol 2000 1994; 5: 112-141.

Ebersole JL, Cappelli D, Sandoval M-N. Subgingival distribution of A.

actinomycetemcomitans in periodontitis. J Clin Periodontol 1994; 21: 65-75.

Ebersole JL, Cappelli D, Sandoval M-N, Steffen MJ. Antigen specificity of serum

antibody in Actinobacillus actinomycetemcomitans infected periodontitis patients. J

Dent Res 1995; 74: 658-666.

Egelberg J. Permeability of the dento-gingival blood vessels. I. Application of the

vascular labelling method and gingival fluid measurements. J Periodontal Res 1966;

1: 180-191.

Egelberg J. The topograph and permeability of vessels at the dento-gingival junction

in dogs. J Periodontal Res 1967; 2: suppi 1.

Egli C, Keung Leung W, Muller K-H, Hancock REW McBride BC. Pore-forming

properties of the major 53-kilodalton surface antigen from the outersheath of

Treponema denticola. Infect Immun 1993; 61: 1694-1699.

Eiselt P. Gingival hypertrophy during pregnancy. Med Jahr Ost Staat 1840; 21: 560.

298

El Attar TM. Prostaglandin E2 in human gingiva in health and disease and its

stimulation by female sex steroids. Prostaglandins 1976; 11: 332-341.

Eley BM Cox SW. Advances in periodontal diagnosis 5. Potential inflammatory and

immune markers. Br Dent J 1998; 184: 220-223.

Eliasson S, Lavstedt S, Ljungheimer C. Radiographic study of alveolar bone height

related to tooth and tooth length. Community Dent Oral Epidemiol 1986; 14: 169-171.

Ellen RP, Song M, Buivids IA. Inhibition of Actinomyces viscosus-Porphyromonas

gingivalis coadhesion by trypsin and other proteins. Oral Microbiol Immunol 1992;

7: 198-203.

Engel D, Monzingo S, Rabinovitch P, Clagett J, Stone R. Mitogen-induced

hyperproliferation response of the peripheral blood mononuclear cells from patients

with severe generalised periodontitis: lack of correlation with proportions of T cells

and T cell subsets. Clin Immunol Immunopath 1984; 30: 374-386.

Engelberger T, Hefti A, Kallenberger A, Rateitschak K-H. Correlations among pailla

bleeding index, other clinical indices and histologically determined inflammation of

gingival papilla. J Clin Periodontol 1983; 10: 579-589.

Ervasti T, Knuuttila M, Pohjamo L, Haukipuro K. Relation between control of

diabetes and gingival bleeding. J Periodontol 1985; 56: 154-157.

Esterberg HL, White PH. Sodium dilantin gingival hyperplasia. J Am Dent Assoc

1945; 32: 16-24.

Farida R, Marsh PD, Newman HN, Rule DC, Ivanyi L. Serological investigation of

various forms of inflammatory periodontitis. J Periodontal Res 1986a; 21: 365-374.

Farida R, Wilson M, Ivanyi L. Serum IgG antibodies to lipopolysaccharides in various

forms of periodontal disease in man. Arch Oral Biol 1986b; 31: 711-715.

299

Fawcett J, Holness CLL, Needham LA, et al. Molecular cloning of ICAM-3, a third

ligand for LFA-1, constitutively expressed on resting leukocytes. Nature 1992; 360:

481-484.

Feldman R, Bravacos J, Rose C. Association between smoking different tobacco

products and periodontal indexes. J Periodontol 1983; 54: 481-487.

Feuille F, Ebersole JL, Kesavalu L, Steffen MJ, Holt SC. Mixed infection with

Porphyromonas gingivalis and Fusobacterium nucleatum in a murine lesion model:

Potential synergistic effects on virulence. Infect Immun 1996; 64: 2095-2100.

Flemmig TF, Selmair I, Schmidt H, Karch H. Specific antibody reactivity against a

110-Kilodalton Actinobacillus actinomycetemcomitans protein in subjects with

periodontitis. Clin Diag Lab Immunol 1996; 3: 678-681.

Flemmig TF. Periodontitis. Ann Periodontol 1999; 4: 32-37.

Fortune F, Walker J, Lehner T. The expression of y8 T cell receptor and prevalence of

primed, activated and IgA bound T cells in Behcet's syndrome. Clin Exp Immunol

1990; 82: 326-332.

Frade R, Barel M, Ehlin-Henriksson B, Klein G. gp140, the C3d receptor of human B

lymphocytes, is also the Epstein-Barr virus receptor. Proc Natl Acad Sci USA 1985;

82: 1490-1493.

Freemont AJ. Adhesion molecules. J Clin Pathol Mol Pathol 1998; 51: 175-184.

Freitas AA, Rocha B, Coutinho AA. Selection of antibody repertoires by anti

idiotypes can occur at multiple steps of B cell differentiation. J Immunol 1986; 136:

466-469.

300

Freudenberg MA, Fornsgaard A, Mitov I Galanos C. ELISA for antibodies to lipid A,

lipopolysaccharides and other hydrophobic antigens. Infection 1989; 17: 322-329.

Ftis A, Singh G, Dolby AE. Antibody to collagen type I in periodontal disease. J

Periodontol 1986; 57: 693-698.

Fujihashi K, Kono Y, Beagley KW, et al. Cytokines and periodontal disease:

Immunopathological role of interleukins for B cell responses in chronic inflamed

gingival tissues. J Periodontol 1993; 64: 400-406.

Fujihashi K, Yamamoto M, Hiroi T, Bamberg TV, McGhee JR, Kiyono H. Selected

Thl and Th2 cytokine mRNA expression by CD4 (+) T cells isolated from inflamed

human gingival tissues. Clin Exp Immunol 1996; 103: 422-428.

Fujita S, Takahashi H, Okabe H, Ozaki Y, Hara Y Kato I. Distribution of natural killer

cells in periodontal diseases: an immunohistochemical study. J Periodontol 1992; 63:

686-689.

Gajewski TF, Fitch FW. Differential activation of murine THI and TH2 clones. Res

Immunol 1991; 142: 19-23.

Galdiero F, Cipollaro de L'Ero G, Bendetto N, Galdiero M, Tufano MA. Release of

cytokines induced by Salmonella typhimurium porins. Infect Immun 1993; 61: 155-

161.

Gapin L. Bravo de Alba Y, Casrouge A, Cabaniols JP, Kourilsky P, Kanellopoulos J.

Antigen presentation by dendritic cells focuses T cell responses against

immunodominant peptides: studies in the hen egg-white lysozyme (HEL) model. J

Immunol 1998; 160: 1555-1564.

Garant PR, Mulvihill JE. The fine structure of gingivitis in the beagle. III. Plasma cell

infiltration of the subepithelial connective tissue. J Periodontal Res 1972; 7: 161-172.

301

Garant PR. Plaque-neutrophil interaction in mono-infected rats as visualised by

transmission electron microscopy. J Periodontol 1976; 47: 132-137.

Garrison S, Holt S, Nichols F. Lipopolysaccharide-stimulated PGE2 release from

human monocytes. J Periodontol 1988; 59: 684-687.

Garrison W, Nichols C. LPS elicited secretory response in monocytes: altered release

of PGE2 but not IL-I beta in patients with adult periodontitis. J Periodontal Res 1989;

24: 88-95

Gemmel E, Feldner B, Seymour GJ. CD45RA and CD45RA positive CD4 cells in

human peripheral blood and periodontal disease tissue before and after stimulation

with periodontopathic bacteria. Oral Microbiol Immunol 1992; 7: 84-88.

Gemmel E, Walsh U, Savage NW. Gemmel GJ. Adhesion molecule expression in

chronic inflammatory periodontal disease tissue. J Periodontal Res 1994; 29: 46-53.

Gemmel E, Seymour GJ. Gamma-delta T lymphocytes in human periodontal disease

tissue. J Periodontol 1995; 66: 780-785.

Gemmel E, Marshall RI, Seymour GJ. Cytokines and prostaglandins in immune

homeostasis and tissue destruction in periodontal disease. Periodontol 2000 1997; 14:

112-143.

Gemmel E, Seymour GJ. Cytokine profiles of cells extracted from humans with

periodontal diseases. J Dent Res 1998; 77: 16-26.

Genco RJ, Evans RT, Ellison SA. Dental research in microbiology with emphasis on

periodontal disease. J Am Dent Assoc 1969; 78: 1016-1036.

Genco RJ, Mashimo PA, Krygier G, Ellison SA. Antibody-mediated effects on the

periodontium. J Periodontol 1974; 45: 330-337.

302

Genco RJ, Taichman NA, Sadowski CA. Precipitating antibodies to Actinobacillus

actinomycetemcomitans in localized juvenile periodontitis. J Dent Res 1980a; 59:

329-336.

Genco RJ, Slots J, Mouton C, Murray P. Systemic immune responses to oral

anaerobic organisms. In: Lambe DW Jr, Genco RJ, Mayberry-Carson KJ, editors.

Anaerobic bacteria: Selected topics. New York: Plenum Publishing Corp., 1980b:

277-293.

Genco RJ, Lee J-Y, Sojar HT, Bedi GS, Loos BG, Dyer DW. Antigenic heterogeneity

of periodontal pathogens. In: Hamada S, Holt S, McGhee J, editors. Periodontal

Disease: Pathogens and Host Immune Responses. Tokyo: Quintessence Publishing

Co. Ltd., 1991: 167-183.

Genco RJ, Sojar H, Lee JY, et al. In: Genco RJ, Hamada S, Lehner T, McGhee J

Mergenhagen S, editors. Molecular pathogenesis of periodontal disease. Washington

DC: ASM Press, 1994: 13-23.

Georgopoulos C. The emergence of the chaperone machines. Trends Biochem Sci

1992; 17: 295-299.

Gibbons RJ, Hay DJ, Schlesinger DH. Delineation of a segment of adsorbed salivary

acidic proline-rich proteins which proteins which promotes adhesion of Streptococcus

gordonii to apatitic surfaces. Infect Immun 1991; 59: 2948-2954.

Gibbs CH, Hirschfield JW, Lee JG, et al. Description and clinical evaluation of a new

computerised periodontal probe. J Clin Periodontol 1988; 15: 137-144.

Gibson WA, Shannon IL. Micro-organisms in human gingival tissues. Periodontics

1964; 2: 119-121.

303

Gigliotti F, Smith L, Insel RA. Reproducible production of human monoclonal

antibodies by fusion of peripheral blood lymphocytes with a mouse myeloma cell line.

J Infect Dis 1984; 149: 43-47.

Gillett R, Cruchley A, Johnson NW. The nature of the inflammatory infiltrates in

childhood gingivitis, juvenile periodontitis and adult periodontitis:

Immunocytochemical studies using monoclonal antibody to HLA Dr. J Clin

Periodontol 1986; 13: 281-288.

Gilsdorf JR, Forney U, LiPuma JJ. Reactivity of Haemophilus influenzae type b anti-

pili antibodies. Microbiol Pathol 1989; 7: 311-316.

Glantz SA. Primer of biostatistics. New York: McGraw-Hill, 1981.

Glassy MC. Production methods for generating human monoclonal antibodies. Hum

Antibod Hybrid 1993; 123: 154-165.

Gmür R, Hrodek K, Saxer UP, Guggenheim B. Doubleblind analysis of the relation

between adult periodontitis and systemic host response to suspected periodontal

pathogens. Infect Immun 1986; 52: 768-776.

Gmür R. Applicability of monoclonal antibodies to quantitiatively monitor

subgingival plaque for specific bacteria. Oral Microbiol Immunol 1988; 3: 187-191.

Gmür R, Strub JR, Guggenheim B. Prevalence of Bacteroides forsythus and

Bacteroides gingivalis in subgingival plaque of prosthodontically treated pateints on

short recall. J Periodontal Res 1989; 24: 113-120.

Gmür R, Guggenheim B. Monoclonal antibodies for the detection of periodontopathic

bacteria. Arch Oral Biol 1990; 35: 145S-151 S.

304

Gmür R, Guggenheim B. Interdental supragingival plaque. A natural habitat of Actinobacillus actinomycetemcomitans, Bacteroides forsythus, Camphylobacter

rectus and Prevotella nigrescens. J Dent Res 1994; 73: 1421-1428.

Gmür R, Baehni PC. Serum immunoglobulin G responses to various Actinohacillus

actinomycetemcomitans serotypes in a young ethnographically heterogeneous

periodontitis patient group. Oral Microbiol Immunol 1997; 12: 1-10.

Goldhaber P. Tissue culture studies of bone as a model system for periodontal

research. J Dent Res 1971; 50: 278-285.

Goldhaber P, Rabadjija L, Beyer WR, Kornhauser A. Bone resorption in tissue culture

and its relevance to periodontal disease. J Am Dent Assoc 1973; 87: 1027-1033.

Golub ES, Mishell RI, Weigle WO, Dutton RW. A modification of the hemolytic

plaque assay for use with protein antigens. J Immunol 1968; 100: 133-137.

Gomes BD, Hausmann E, Wienfeld N, De Luca C. Prostaglandins: bone resorption

stimulating factors released from monkey gingiva. Calcif Tissue Res 1976; 19 : 285-

293.

Gordon J, Graeme G. Control of human B-lymphocyte growth and differentiation. In:

Bird G, Calvert JE, editors. B lymphocytes in human disease. Oxford, England:

Oxford University press, 1988: 148-172.

Gottlieb B. Tissues changes in pyorrhea. J Am Dent Assoc 1927; 4: 2178-2207.

Gray D. Immunological memory. Annu Rev Immunol 1993; 11: 49-77.

Greenstein G, Caton J, Poison AM. Histologic characteristics associated with bleeding

after probing and visual signs of inflammation. J Periodontol 1981; 52: 420-425.

305

Gregory RL, Kim DE, Kindle JC, Hobbs LC, Lloyd DR. Immunoglobulin-degrading

enzymes in localized juvenile periodontitis. J Periodontal Res 1992; 27: 176-183.

Grenier D, Uitto V-J, McBride BC. Cellular location of a Treponema denticola

chymotrypsin-like protease and importance of the protease in migration through the

basement membrane. Infect Immun 1990; 58: 347-351.

Grenier D. Antagonistic effect of oral bacteria towards Treponema denticola. J Clin

Microbiol 1996: 34: 1249-1252.

Grey HM, Abel CA, Yount WJ, Kunkel HG. A subclass of human yA-globulins (yA2)

which lacks the disulfide bonds linking heavy and light chains. J Exp Med 1968; 128:

1223-1236.

Griffiss JM. Mechanisms of Host Immunity. In: Cartwright K, editor. Meningococcal

Disease. Chichester: John Wiley & Sons, 1995: 35.

Grossi S, Zambon J, Ho A, et al. Assessment of risk for periodontal disease. I. Risk

indicators for attachment loss. J Periodontol 1994; 65: 260-267.

Grossi S, Genco R, Machtei E, et al. Assessment of risk for periodontal disease. II.

Risk indicators for alveolar bone loss. J Periodontol 1995; 66: 23-29.

Gunsolley JC, Tew JG, Gooss CM, Burmeister JA, Schenkein HA. Effects of race and

periodontal status on antibody reactive with Actinobacillus actinomycetemcomitans

strain Y4. J Periodontal Res 1988; 23: 303-307.

Gunsolley JC, Tew JG, Connor T, Burmeister JA, Schenkein HA. Relationship

between race and antibody reactive with periodontitis-associated bacteria. J


306

Gusberti FA, Syed SA, Bacon G, Grossman N, Loesche WJ. Puberty gingivitis in

insulin-dependent diabetic children. I. Cross-sectional observations. J Periodontol

1983; 54: 714-720.

Haapasalo M, Muller KH, Uitto V-J, Leung WK McBride BC. Characterisation,

cloning and binding properties of the major 53-kilodalton Treponema denticola

surface antigen. Infect Immun 1992; 60: 2058-2065.

Haas W, Kaufman S, Martinez-A C. The development and function of y8 T cells.

Immunol Today 1990; 11: 340-343.

Haber J, Kent RL. Cigarette smoking in periodontal practice. J Periodontol 1992; 63:

100-106.

Haber J, Wattles J, Crowley M, Mandell R, Joshipura K, Kent RL. Evidence for

cigarette smoking as a major risk factor for periodontitis. J Periodontol 1993; 64: 16-

23.

Haber J. Smoking is a major risk factor for periodontitis. Curr Opin Periodontol 1994;

12-18.

Haffajee AD, Socransky SS, Ebersole JL, Smith DJ. Clinical, microbiological and

immunological features associated with the treatement of active periodontosis lesions.

J Clin Periodontol 1984; 11: 600-618.

Haffajee AD, Socransky SS, Smith C, Dibart S. Relation of baseline microbial

parameters to future periodontal attachment loss. J Clin Periodontol 1991; 18: 744-

750.

Haffajee AD, Socransky SS, Goodson JM. Subgingival temperature (I) Relation to baseline clinical parameters. J Clin Periodontol 1992; 19: 401-408.

307

Haffajee AD, Socransky SS. Microbial etiological agents of destructive periodontal diseases. Periodontol 2000 1994; 5: 78-111.

Haffajee AD, Dibart S, Kent Jr RL, Socransky SS. Factors associated with different

responses to periodontal therapy. J Clin Periodontol 1995; 22: 628-636.

Haffajee AD, Cugini MA, Dibart S, Smith C, Kent Jr RL, Socransky SS. The effect of SRP on the clinical and microbiological parameters of periodontal diseases. J Clin

Periodontol 1997; 24: 324-334.

Hagiwara E, Gourley MF, Lee S, Klinman DK. Disease severity in patients with

systemic lupus erythematosus correlates with an increased ratio of interleukin-10:

interferon-gamma-secreting cells in the peripheral blood. Arthritis Rheum 1996; 39:

379-385.

Halliwell B, Gutteridge JMC. The importance of free radicals and catalytic metal ions

in human disease. Mol Aspects Med 1985; 8: 89-193.

Hamada N, Watanabe K, Sasakawa C, Yoshikawa M, Yoshimura F Umemoto T.

Construction and characterisation of a FimA mutant of Porphyromonas gingivalis.

Infect Immun 1994; 62: 1696-1704.

Hanazawa S, Amano S, Nakda K, et al. Biological characterisation of interleukin-1

like cytokine product produced by cultured bone cells from newborn mouse calvaria.

Calcif Tissue Int 1987; 141: 31-37.

Hanna MG, Francis MW, Peters LC. Localization of 125 I-labelled antigen in

germinal centers of mouse spleen: effects of competitive infection of specific or non-

cross-reacting antigen. Immunology 1968; 15: 75-91.

Hart TC, Marazita ML, McCanna KM, Schenkein HA, Diehl SR. Re-evaluation of the

chromosome 4q candidate region for early onset periodontitis. Hum Genet 1993; 91:

416-422.

308

Hart TC. Genetic considerations of risk in human periodontal disease. Curr Opin

Periodontol 1994; 3-11.

Hart TC. Genetic risk factors for early-onset periodontitis. J Periodontol 1996; 67:

355-366.

Hart TC, Kornman KS. Genetic factors in the pathogenesis of periodontitis.

Periodontol 2000 1997; 14: 202-215.

Hassell TM. Phenytoin: gingival overgrowth. In: Myers HM, editor. Epilepsy and the

oral manifestations of phenytoin therapy, vol. 9. Basel: S. Karger AG, 1981: 116-202.

Hassell TM, Hefti AF. Drug-induced gingival overgrowth: old problem, new problem.

Crit Rev Oral Biol Med 1991; 2: 103-137.

Hassell T. Tissues and cells of the periodontium. Periodontology 2000 1993; 3: 9-38.

Haston WS, Shields JM. Contraction waves in lymphocyte locomotion. J Cell Sci

1984; 68: 227-241.

Haubek D, Poulsen K, Asikainen S, Kilians M. Evidence for absence in Northern

Europeans of especially virulent clonal types of Actinobacillus

actinomycetemcomitans. J Clin Microbiol 1995; 33: 395-401.

Hausmann E, Ralsz LG, Muller WA. Endotoxin: Stimulation of bone resorption tissue

culture. Science 1970; 168: 862-864.

Havarth L. Neutrophil chemotactic factors. In: Goldberg ID, editor. Cell motility

factors. Basel Birkhauser Verlag, 1991

Hefti AF, Eshenaur AE, Hassell TM, Stone C. Gingival overgrowth in cyclosporin A

treated multiple sclerosis patients. J Periodontol 1994; 65: 744-749.

309

Heremans JF. Immunoglobulin A. In: Sela M, editor. The antigens. vol 2. New York:

Academic press inc., 1974: 365.

Herr W, Linn B, Leister N, Wandel E, Meyer zum Buschenfelde KH, Wolfel T. The

use of computer-assisted video image analysis for the quantification of CD8+ T

lymphocytes producing tumor necrosis factor alpha spots in response to peptide

antigens. J Immunol Methods 1997; 203: 141-152.

Hillmann G, Hillmann B, Geurtsen W. Immunohistological determination of

interleukin-1 beta in inflamed human gingival epithelium. Arch Oral Biol 1995; 40:

353-359.

Hinode D, Grenier D, Mayrand D. A general procedure for the isolation of heat shock

proteins from periodontopathogenic bacteria. J Microbiol Methods 1996; 25: 349-355.

Hinode D, Nakamura R, Grenier D, Mayrand D. Cross reactivity of specific antibodies

directed to heat shock proteins from periodonto-pathogenic bacteria of human origin.

Oral Microbiol Immunol 1998; 13: 55-58.

Hirano T. Interleukin-6. In Thomson A, editor. The cytokine Handbook, edition 1.

New York: Academic Press, 1991: 169-190.

Hirsch HZ, Tarkowski A, Miller EJ, Gay S, Koopman WJ, Mestecky J. Autoimmunity

to collagen in adult periodontal disease. J Oral Pathol 1988; 17: 456-459.

Holm-Pederson P, Agerbaek N, Theilade E. Experimental gingivitis in young and

elderly individuals. J Clin Periodontol 1975; 2: 14-24.

Holst 0, Brade H. In: Morrison DC, Ryan JL, editors. Bacterial Endotoxic

Lipopolysaccharides: Chemical Structure of the Core Region of Lipopolysaccharides.

Boca Raton, FL: CRC Press, 1992: 135.

310

Holt SC, Bramanti TE. Factors in virulence expression and their role in periodontal

disease pathogenesis. Crit Rev Oral Biol Med 1991; 2: 177-281.

Holt SC, Ebersole JL. The surface of selected periodontopathic bacteria: Possible role

in virulence. In: Hamada S, Holt SC, McGhee JR, editors. Periodontal Disease

Pathogens and Host Immune Responses. Tokyo: Quintessence publishing co. Ltd.,

1991: 79-98.

Holt SC, Kesavalu L, Walker S, Genco CA. Virulence factors of Porphyromonas

gingivalis. Periodontol 2000 1999; 20: 168-238.

Hopewell-Smith A. Pyorrhea alveolaris - its pathohistology (1). Preliminary note (2).

Concluding remarks. Dental Cosmos 1911; 53: 387-411.

Horning GM, Hatch CL, Cohen ME. Risk indicators for periodontitis in a military

treatment population. J Periodontol 1992; 63: 297-302.

Horibe M, Watanabe H, Ishikawa I. Effect of periodontal treatments on serum IgG

antibody titers against periodontopathic bacteria. J Clin Periodontol 1995; 22: 510-

515.

Horton JE, Raisz LG, Simmons HA, Oppenheim JJ, Mergenhagen SE. Bone resorbing

activity in supernatant fluid from cultured human peripheral blood leukocytes. Science

1972; 177: 793-795.

Howe DK, Crawford AC, Lindsay D, Sibley LD. The p29 and p35 immunodominant

antigens of Neospora caninum tachyzoites are homologous to the family of surface

antigens of Toxoplasma gondii. Infect Immun 1998; 66: 5322-5328.

311

Howells G L. Cytokine networks in destructive periodontal disease. Oral Diseases

1995; 1: 266-270.

Hsu SD, Cisar JO, Sandberg AL, Kilian M. Adhesive properties of viridans

streptococcal species. Microb Ecol Health Dis 1994; 7: 125-137.

Hugoson A. Gingivitis in pregnant women. A longitudinal clinical study.

Odontologisk Revy 1971; 22: 65-84.

Hugoson A, Thorstensson H, Falk H, Kuylenstierna J. Periodontal conditions in

insulin-dependent diabetics. J Clin Periodontol 1989; 16: 215-223.

Inderlied CB, Kemper CA, Bermudez LEM. The Mycobacterium avium complex.

Clin Microbiol Rev 1993; 6: 266-310.

Ishihara Y, Nishihara T, Maki E, Noguchi T, Koga T. Role of interleukin-1 and

prostaglandin in in vitro bone resorption induced by Actinobacillus

actinomycetemcomitans lipopolysaccharide. J Periodontal Res 1991; 26: 155-160.

Ishikawa I, Nakashima K, Koseki T, et al. Induction of the immune response to

periodontopathic bacteria and its role in the pathogenesis of periodontitis. Periodontol

2000 1997; 14: 79-111.

Ismaiel MO, Greenman J, Scully C. Serum antibodies against the trypsin-like protease

of Bacteroides gingivalis in periodontitis. J Periodontal Res 1988; 23: 193-198.

Jacob J, Kelsoe G, Rajewsky K, Weiss U. Intraclonal generation of antibody mutants

in germinal centers. Nature 1991a; 354: 389-392.

Jacob J, Kassir R, Kelsoe G. In situ studies of the primary immune response to (4-

hydroxy-3-nitrophenyl) acetyl. I. The architecture and dynamics of responding cell

populations. J Exp Med 1991b; 173: 1165-1175.

312

Jacob J, Kelsoe G. In situ studies of the primary immune response to (4-hydroxy-3-

nitrophenyl) acetyl. II. A common clonal origin for periarteriolar lymphoid sheath-

associated foci and germinal centers. J Exp Med 1992; 176: 679-687.

Jacobson EB, Thorbecke GJ. Relationship of germinal centers in lymphoid tissue to

immunologic memory. IV. Formation of 19S and 7S antibody by splenic white and

red pulp during the secondary response in vitro. Lab Invest 1968; 19: 635-642.

James K, Bell GT. Human monoclonal antibody production. Current status and future

prospects. J Immunol Methods 1987; 100: 5-40.

James WW, Counsel A. A histologic investigation into "socalled pyorrhea alveolaris".

Br Dent J 1927; 48: 1237-1251.

Jandinski JJ, Stashenko P, Feder LS, et al. Localization of interleukin-1 beta in human

periodontal tissue. J Periodontol 1991; 62: 36-43.

Janeway CA, Travers P. The immune system in health and disease. Immunobiology.

Third edition. Oxford: Blackwell, 1994.

Jarjour WN, Jeffries BD, Davis JS, Welch WJ, Mimura T, Winfield JB.

Autoantibodies to human stress proteins. Arthritis Rheum 1991; 34: 1133-1138.

Jenkins WMM, Kinane DF. The `high risk' group in periodontitis. Br Dent J 1989;

167: 168-171.

Jerne NK, Nordin AA. Plaque formation in agar by single antibody-producing cells.

Science 1963; 140: 405.

Jerne NK, Henry C, Nordin AA, Fuji H, Koros AMC, Leifkovits I. Plaque forming

cells: methodology and theory. Transplant Rev 1974; 18: 130-191.

313

Johnson RT, Rao PN. Mammalian cell fusion: Induction of premature chromosome

condensation in interphase nuclei. Nature 1970; 226: 717-722.

Kagan JM. Local immunity to Bacteroides gingivalis is periodontal disease. J Dent

Res 1980; 59: 1750-1756.

Kaila M, Isolauri E, Virtanen E, Arvilommi H. Preponderance of IgM from blood

lymphocytes in response to infantile rotavirus gastroenteritis. Gut 1992; 33: 639-642.

Kamagata Y, Miyasaka N, Inoue H, Hashimoto J, lida M. Cytokine production in

inflamed human gingival tissues, interleukin-6. J Jpn Assoc Periodontol 1989; 31:

1081-1087.

Kantele AM, Takanen R, Arvilommi H. Immune response to acute diarrhea seen as

circulating antibody-secreting cells. J Infect Dis 1988; 158: 1011-1016.

Kantele A, Mäkelä PH. Different profiles of the human immune response to primary

and secondary immunization with an oral Salmonella typhi Ty2la vaccine. Vaccine

1991; 9: 423-427.

Kantele A, Papunen R, Virtanen E, et al. Antibody-secreting cells in acute urinary

tract infection as indicators of local immune response. J Infect Dis 1994; 169: 1023-

1028.

Kaplan SL, Mason EO Jr, Wiederuran BL. Role of adherence in the pathogenesis of

Haemophilus influenzae type b infection in infant rats. Infect Immun 1983; 42: 612-

617.

Kappler J, Kotzin B, Herron L, Gelfand EW, Bigler RD, Boylston A. Vß specific

stimulation of human T cells by staphylococcal toxins. Science 1989; 244: 811-813.

314

Karttunen R, Karttunen T, Ekre HP, MacDonald TT. Interferon gamma and interleukin 4 secreting cells in the gastric antrum in Heliobacter pylori positive and

negative gastritis. Gut 1995; 36: 341-345.

Katende J, Morzaria S, Toye P, et al. An enzyme-linked immunosorbent assay for

detection of Theileria parva antibodies in cattle using a recombinant polymorphic immunodominant molecule. Parasitol Res 1998; 84: 408-416.

Kato S, Muro M, Akifusa S, et al. Evidence for apoptosis of murine macrophages by

Actinobacillus actinomycetemcomitans infection. Infect Immun 1995; 63: 3914-3919.

Katz J, Benoliel R, Goultschin J, Brautbar C. Human leukocyte antigen (HLA) DRY

positive association with rapidly progressive periodontitis. J Periodontol 1987; 58:

607-610.

Katz J, Goultschin J, Benoliel R, Schlesinger M. Peripheral T lymphocyte subsets in

rapidly progressive periodontitis. J Clin Periodontol 1988; 15: 266-268.

Kaufman AY. An oral contraceptive as an etiologic factor in producing hyperplastic

gingivitis and a neoplasm of the pregnancy tumor type. Oral Surg Oral Med Oral

Pathol 1969; 28: 666-670.

Kaufman SHE. Heat Shock Proteins and the Immune Response. Immunol Today

1990; 11: 129-136.

Kawahara K, Fukunga M, Takata T, Kawamura K, Morishita M, Iwamoto Y.

Immunohistochemical study of gamma-delta T cells in human gingival tissues. J

Periodontol 1995; 66: 775-779.

Kawai T, Ito HO, Sakato N, Okado H. A novel approach for detecting an immunodominant antigen of Porphyromonas gingivalis in diagnosis of adult

periodontitis. Clin Diagn Lab Immunol 1998; 5: 11-17.

315

Keller HU, Zimmermann A, Cottier H. Crawling-like movements, adhesion to solid

substrata and chemokinesis of neutrophil granulocytes. J Cell Sci 1983; 64: 89-106.

Kelsoe G. The germinal center: a crucible for lymphocyte selection. Semin Immunol

1996; 8: 179-184.

Kesavalu L, Holt SC, Ebersole JL. Hydrolytic enzymes of Porphyromonas gingivalis:

In vitro distribution and functions. Int J Oral Biol 1997; 22: 255-267.

Kesavalu L, Holt SC, Ebersole JL. Virulence of a polymicrobic complex. Treponenia

denticola and Porphyromonas gingivalis in a murine model. Oral Microbiol Immunol

1998; 13: 373-377.

Kilian M. Degradation of immunoglobulin Al, A2 and G by suspected principal

periodontal pathogens. Infect Immun 1981; 34: 757-764.

Kinane DF, Cullen CF, Johnston FA, Evans CW. Neutrophil chemotactic behaviour in

patients with early-onset forms of periodontitis. I. Leading front analysis in Boyden

chambers. J Clin Periodontol 1989a; 16: 242-246.

Kinane DF, Johnstone FA, Evans CW. Depressed helper-to-surpressor T cell ratios in

early onset forms of periodontal disease. J Periodontal Res 1989b; 24: 161-164.

Kinane DF, Davies RM. Periodontal manifestations of systemic disease. In: Jones JH,

Mason DK, editors. Oral manifestations of systemic disease. London: Bailliere-

Tindall, 1990: 512-536.

Kinane DF, Adonogianaki E, Moughal N, Winstanley FP, Mooney J, Thornhill M.

Immunocytochemical characterization of cellular infiltrate, related endothelial

changes and determination of GCF acute-phase proteins during human experimental

gingivitis. J Periodontal Res 1991; 26: 286-288.

316

Kinane DF, Mooney J, MacFarlane TW, McDonald M. Local and systemic antibody

response to putative periodontopathogens in patients with chronic periodontitis:

correlation with clinical indices. Oral Microbiol Immunol 1993; 8: 65-68.

Kinane DF, Lindhe J. Pathogenesis of plaque associated periodontal disease. In:

Lindhe L, editor. Textbook of periodontology. Copenhagen, Denmark: Munksgaard,

1997.

Kinane DF, Mooney J, Ebersole JL. Humoral immune response to Actinohacillus

actinomycetemcomitans and Porphyromonas gingivalis in periodontal disease.

Periodontol 2000 1999a; 20: 289-340.

Kinane DF, Lappin DF, Koulouri 0, Buckley A. Humoral immune responses in

periodontal disease may have mucosal and systemic immune features. Clin Exp

Immunol 1999b; 115: 534-541.

Klein DC, Raisz LG. Prostaglandins: stimulation of bone resorption in tissue culture.

Endocrinology 1970; 86: 1436-1440.

Koga T, Nishihara T, Fujiwara T, et al. Biochemical and immunobiological properties

of lipopolysaccharide (LPS) from Bacteroides gingivalis: comparison with LPS from

Escherichia coli. Infect Immun 1984; 47: 638-647.

Köhler G, Milstein C. Continuous cultures of fused cells secreting antibody of

predefined specificity. Nature 1975; 256: 495-497.

Kopp W. Density and localisation of lymphocytes with natural killer (NK) cell activity

in periodontal biopsy specimens from patients with severe periodontitis. J Clin

Periodontol 1988; 15: 595-600.

Komman KS, Crane A, Wang HY et al. The interleukin 1 genotype as a severity

factor in adult periodontal disease. J Clin Periodontol 1997; 24: 72-77.

317

Korostoff J, Wang JF, Kieba I, Miller M, Shenker BJ, Lally ET. Actinobacillus

actinomycetemcomitans leukotoxin induces apoptosis in HL-60 cells. Infect Immun

1998; 66 : 4474-4483.

Koshland ME. Structure and function of the J chain. Adv Immunol 1975; 20: 41-69.

Kozbor D, Lagarde AE, Roder JC. Human hybridomas constructed with antigen-

specific Epstein-Barr virus-transformed cell lines. Proc Natl Acad Sci USA 1982; 79:

6651-6655.

Kraal G, Weissman L, Butcher EC. Germinal centre B cells: antigen specificity and

changes in heavy chain class expression. Nature 1982; 298: 377-379.

Kristofferson T, Tonder G. Anti-immunoglobulin activity in inflamed human gingiva.

J Dent Res 1973; 52: 991 abstr.

Kroese FGM, Timens W, Nieuwenhuis P. Germinal center reaction and B

lymphocytes: Morphology and function. Curr Top Pathol 1990; 84: 103-148.

Krogfelt KA. Bacterial adhesion: genetics, biogenesis and role in pathogenesis of

fimbrial adhesins of Escherichia coli. Rev Infect Dis 1991; 13: 721-735.

Kulkarni GV, Chan KH, Sandham HJ. An investigation into the use of restriction

endonuclease analysis for the study of transmission of mutans streptococci. J Dent Res

1989; 86: 1155-1161.

Kusumoto Y, Ogawa T, Hamada S. Generation of specific antibody-secreting cells in

salivary glands of BALB/c mice following parental or oral immunization with

Porphyromonas gingivalis fimbriae. Arch Oral Biol 1993; 38: 361-367.

Lai CH, Listgarten MA, Shirakawa M, Slots J. Bacteroides forsythus in adult

gingivitis and periodontitis. Oral Microbiol Immunol 1987; 2: 152-157.

318

Lala A, Anamo A, Sojar HT, Radel SJ, De Nardin E. Porphyromonas gingivalis

trypsin-like protease: a possible natural ligand for the neutrophil formyl peptide

receptor. Biochem Biophys Res Commun 1994; 199: 1489-1496.

Lally ET, Kieba IR, Golub EE, Lear JD, Tanaka JC. Structure/function of

Actinobacillus actinomycetemcomitans leukotoxin. J Periodontol 1996; 67: 298-308.

Lane RD, Crissman RS, Lachman MF. Comparison of polyethylene glycols as

fusogens for producing lymphocyte myeloma hybrids. J Immunol Methods 1984; 72:

71-76.

Lantz MS, Allen RD, Duck LW, Blume JL, Switalski LM, Hook M. Identification of

Porphyromonas gingivalis Components that Mediate its Interactions with Fibronectin.

J Bacteriol 1991a; 173: 4263-4270.

Lantz MS, Allen RD, Vail TA, Switalski LM, Hook M. Specific Cell Components of

Bacteroides gingivalis Mediate Binding and Degradation of Human Fibrinogen. J

Bacteriol 1991b; 173: 495-501.

Lappin DF, Koulouri 0, Radvar M, Hodge P, Kinane DF, Relative proportions of

mononuclear cell types in periodontal lesions analysed by immunohistochemistry. J

Clin Periodontol 1999; 26: 183-189.

Larsen E, Palabrica T, Sajer S, et al. PADGEM-Dependent adhesion of platelets to

monocytes and neutrophils is mediated by a lineage-specific carbohydrate, LNF III

(CD 15). Cell 1990; 63: 467-474

Last J. A dictionary of epidemilogy. 2 "d edition. New York: Oxford University Press,

1988: 115-116.

Laver WG, Air MN, Webster RG, Smith-Gill SJ. Epitopes on protein antigens: Misconceptions and realities. Cell 1990; 61: 553-556.

319

Lee FK, Nahmias AJ, Lowery S, et al. ELISPOT: a new approach to studying the

dynamics of virus-immune system interaction for diagnosis and monitoring of HIV

infection. AIDS Res Hum Retroviruses 1989; 5: 517-523.

Leung K-P, Fukushima H, Sagawa H, Walker CB, Clark WB. Surface appendages,

haemagglutination and adherence to human epithelial cells of Bacteroides

intermedius. Oral Microbiol Immunol 1989; 4: 204-210.

Leung E, Greene J, Ni J, et al. Cloning of the mucosal addressin MAdCAM-1 from

human brain: identification of novel alternatively spliced transcripts. Immunol Cell

Biol 1996; 74: 490-496.

Liljemark WF, Bloomquist C. Human oral microbial ecology and dental caries and

periodontal diseases. Crit Rev Oral Biol Med 1996; 7: 180-198.

Liljenberg B, Lindhe J, Berglundh T, Dahlen G. Some microbiological,

histopathological and immunohistochemical characteristics of progressive periodontal

disease. J Clin Periodontol 1994; 21: 720-727.

Lindemann RA. Bacterial activation of human natural killer cells: role of cell surface

lipopolysaccharide. Infect Immun 1988; 56: 1301-1308.

Lindhe JJ, Hamp SE, Löe H. Experimental periodontitis in the beagle dog. J


Lindhe J, Rylander H. Experimental gingivitis in young dogs. Scand J Dent Res 1975;

83: 314-326.

Lindhe J, Hamp S-E, Löe H. Plaque-induced periodontal disease in beagle dogs. A 4-

year clinical roentgenographical and histometric study. J Periodontal Res 1975; 10:

243-255.

320

Lindhe J, Liljenberg B, Listgarten M. Some microbiological and histopathological

features of periodontal disease in man. J Periodontol 1980; 51: 264-269.

Lindhe J, Westfelt E, Nyman S, Socransky SS, Heijl J, Bratthall G. Healing following

surgical/nonsurgical treatment of periodontal disease. A clinical study. J Clin

Periodontol 1982; 9: 115-128.

Lindhe J, Karring T. Anatomy of the Periodontium. In: Lindhe J, Karring T, Lang NP,

editors. Clinical periodontology and implant dentistry. Third edition. Copenhagen:

Munksgaard; 1997: 19-68.

Listgarten MA. Electron microscopic study of the gingivo-dental junction of man. Am

J Anat 1966; 119: 147-177.

Listgarten MA, Ellegaard B. Experimental gingivitis in the monkey. Relationship of

leucocyte in junctional epithelium, sulcus depth and connective tissue inflammation

scores. J Periodontal Res 1973; 8: 199-214.

Listgarten MA, Mayo HE, Tremblay R. Development of dental plaque on epoxy resin

crowns in man. A light and electron microscopic study. J Periodontol 1975; 46: 10-26.

Listgarten M, Hellden L. Relative distribution of bacteria at clinically healthy and

periodontally diseases sites in humans. J Clin Periodontol 1978; 5: 115-132.

Listgarten MA, Lai C-H, Evian Cl. Comparative antibody titers to Actinobacillus

actinomycetemcomitans in juvenile periodontitis, chronic periodontitis and

periodontally healthy subjects. J Clin Periodontol 1981; 8: 155-164.

Listgarten MA, Schifter CC, Laster L. 3-year longitudinal study of the periodontal

status of an adult population with gingivitis. J Clin Periodontol 1985; 12: 225-238.

321

Liu Y-J, Johnson GD, Gordon J, MacLennan ICM. Germinal centers in T-cell

dependent antibody responses. Immunol Today 1992; 13: 17-21.

Löe H. Physiological aspects of the gingival pocket. An experimental study. Acta

Odontol Scand 1961; 19: 387-395

Löe H, Silness J. Periodontal disease in preganancy. I. Prevalence and severity. Acta

Odontol Scand 1963; 21: 533-551.

Löe H. Periodontal changes in pregnancy. J Periodontol 1965; 36: 209-216.

Löe H, Theilade E, Jensen SB. Experimental gingivitis in man. J Periodontol 1965;

36: 177-187.

Löe H, Holm-Pedersen P. Absence and presence of fluid from normal and inflamed

gingivae. Periodontics 1965; 3: 171-177.

Loesche WJ. Chemotherapy of dental plaque infections. Oral Sci Rev 1976; 965-107.

Loesche WJ, Syed SA. Bacteriology of human experimental gingivitis: effect of

plaque and gingivitis score. Infect Immun 1978; 21: 830-839.

Loesche WJ, Syed SA, Laughon BE, Stoll J. The bacteriology of acute necrotising

ulcerative gingivitis. J Periodontol 1982; 53: 223-230.

Loesche WJ, Syed SA, Schmidt E, Morrison EC. Bacterial profiles of subgingival

plaques in periodontitis. J Periodontol 1985; 56: 447-456.

Loesche WJ, Lopatin DE, Stoll J, van Poperin N, Hujoel PP. Comparison of various

detection methods for periodontopathic bacteria - Can culture be considered the

primary reference standard? J Clin Microbiol 1992; 30: 370-376.

322

Lo-Man R, Langeveld JP, Martineau P, Hofnung M, Meloen RH, Leclerc C.

Immunodominance does not result from peptide competition for MHC class II

presentation. J Immunol 1998; 160: 1759-1766.

Lu H, Califano JV, Shenkein HA, Tew JG. Immunoglobulin class and subclass

distribution of antibodies reactive with the immunodominant antigen of Actinohacillus

actinomycetemcomitans serotype b. Infect Immun 1993; 61: 2400-2407.

Luderitz 0, Freudenberg MA, Galanos C, Lehmann V, Rietschel ET, Shaw DH.

Lipopolysacchaides of gram negative bacteria. Curr Top Membr Transp 1982; 17: 79-

151.

Luger T, Stadler B, Katz S, Oppenheim J. Epidermal cell (keratinocyte) derived

activating factor. J Immunol 1981; 127: 1493-1498.

Lundqvist C, Hammarström ML. T-cell receptor gamma delta-expressing

intraepithelial lymphocytes are present in normal and chronically inflamed human

gingiva. Immunology 1993; 79: 38-45.

Lundqvist C, Baranov V, Teglund S, Hammarström S, Hammarström ML. Cytokine

profile and ultrastructure of intraepithelial gamma delta T cells in chronically

inflamed human gingiva suggest a cytotoxic effector function. J Immunol 1994; 153:

2302-2312.

Lynch MA, Ship H. Initial oral manifestations of leukaemia. J Am Dent Assoc 1967;

75: 932-940.

Lynn BD. "The pill" as an etiologic agent in hypertrophic gingivitis. Oral Surg Oral

Med Oral Pathol 1969; 24: 333-334.

323

MacDermott RP, Chess L, Schlossuran SF. Immunolgoic functions of isolated human

lymphocyte subpopulation. V. Isolation and funtional analysis of a surface Ig negative,

E rosette negative subset. Clin Immunol Immunolpathol 1975; 4: 415-424.

MacFarlane TW, Jenkins WMM, Gilmour WH, McCourtie D, McKenzie D.

Longitudinal study of untreated periodontitis. (II). Microbiological findings. J Clin

Periodontol 1988; 15: 331-337.

Machtei E, Dunford R, Grossi S, Genco R. Cumulative nature of periodontal

attachment loss. J Periodontal Res 1994; 29: 361-364.

MacLennan ICM, Gray D. Antigen-driven selection of virgin and memory B cells.

Immunol Rev 1986; 91: 61-85.

MacLennan ICM, Liu Y-J, Oldfield S, Lane PJL. The evolution of B cell clones. Curr

Top Microbiol Immunol 1990; 158: 37-63.

MacLennan ICM, Liu Y-J, Johnson GD. Maturation and dispersal of B cell clones

during T cell-dependent antibody response. Immunol Rev 1992; 126: 143-161.

Maclennan IC. Germinal centers. Annu Rev Immunol 1994; 12: 117-139.

Mackler BF, Frostad KB, Robertson RB, Levy BM. Immunoglobulin-bearing

lymphocytes and plasma cells in human periodontal disease. J Periodontal Res 1977;

12: 37-45.

Madianos PN. Lappin D, Kinane DF, Papapanou PN, Sandros J. Porplryromonas

gingivalis induces IL-Iß mRNA expression in oral epithelium. In "Interactions of

Porphyromonas gingivalis with oral epithelial cells (thesis). Univ. of Gothenburg.

1997.

Maeda H, Miyamoto M, Hongyo H, Nagai A, Kurihara H, Murayama Y. Heat shock

protein 60 (Gro EL) from Porphyromonas gingivalis: Cloning and sequence analysis

324

of its gene and purification of the recombinant protein. FEMS Microbiol Lett 1994;

119: 129-136.

Mäkelä M, Nikkari S, Meurman 0, Laine M, Arvilommi H. Virus-specific, antibody-

secreting cells during upper respiratory infections. J Med Virol 1995; 47: 416-420.

Malek R, Fisher G, Caleca A, et al. Inactivation of the Porphyronionas gingivalis

FimA gene blocks periodontal damage in gnotobiotic rats. J Bacteriol 1994; 176:

1052-1059.

Mallison SM III, Kaugars C, Szakal Ak, Shenkein HA, Tew JG. Synthesis of antibody

specific for nonoral antigen in the gingiva of periodontitis patients. J Periodontal Res

1989; 24: 214-216.

Mandell RL. A longitudinal microbiological investigation of Actinohacillus

actinomycetemcomitans and Eikenella corrodens in juvenile periodontitis. Infect

Immun 1984; 45: 778-780.

Mandell RL, Tripodi LS, Savitt E, Goodson JM, Socransky SS. The effect of

treatment on Actinobacillus actinomycetemcomitans in localized juvenile

periodontitis. J Periodontol 1986; 57: 94-99.

Mangan DF, Taichman NS, Lally ET, Wahl SM. Lethal effects of Actinrobacillus

actinomycetemcomitans leukotoxin on human T lymphocytes. Infect Immun 1991; 59:

3267-3272.

Manhart SS, Reinhardt RA, Payne JB, et al. Gingival cell IL-2 and IL-4 in early onset

peridodontitis. J Periodontol 1994; 144: 662-666.

Manogue KR, Van Deventer SJH, Cerami A. A tumor necrosis factor or cachectin. In

Thomson A editor. The Cytokine Handbook. First edition. New York: Academic

Press; 1991: 241-256.

325

Manouchehr-Pour M, Spagnuolo PJ, Bissada NF. Impaired neutrophil chernotaxis in

diabetic patients with severe periodontitis. J Dent Res 1981 a; 60: 729-730.

Manouchehr-Pour M, Spagnuolo PJ, Rodman HF, Bissada NF. Comparison of

neutrophil chemotactic response in diabetic patients with mild and severe periodontal

disease. J Periodontol 1981b; 52: 410-414.

Mansheim BJ, Onderdonk AB, Kasper DL. Immunochernical and biologic studies of

the lipopolysaccharide of Bacteroides melaninogenicus subspecies asaccharolyticus. J

Immunol 1978; 120: 72-78.

Mansheim BJ, Strenstrorn ML, Low SB, Clark WB. Measurement of serum and

salivary antibodies to the oral pathogen Bacteroides asaccharolyticus in human

subjects. Arch Oral Biol 1980; 25: 553-557.

Manthey CL, Vogel SN. Interactions of lipopolysaccharide with macrophages.

Immunol Ser 1994; 60: 63-81.

Marazita ML, Burmeister JA, Gunsolley JC, Koertge TE, Lake K, Schenkein HA.

Evidence for autosomal dominant inheritance and race specific heterogeneity in early

onset periodontitis. J Periodontol 1994; 65: 623-630.

Marazita ML, Lu H, Cooper ME et al. Genetic segregation analyses of serum IgG2

levels. Am J Hum Genet 1996; 58: 1042-1049.

Marlin SD, Springer TA. Purified intercellular adhesion molecule-I (ICAM-1) is a

ligand for lymphocyte function associated lymphocyte antigen-1 (LFA-1). Cell 1987;

51: 813-819.

Mariotti A. Sex steroid hormones and cell dynamics in the periodontium. Crit Rev

Oral Biol Med 1994; 5: 27-53.

326

Mariotti A. Dental plaque-induced gingival diseases. Ann Periodontol 1999; 4: 7-17.

Marsh PD. Host defenses and microbial homeostasis: Role of microbial interactions. J

Dent Res 1989; 68: 1567-1575.

Marsh PD, Bradshaw DJ. Dental plaque as a biofilm. J Ind Microbiol 1995; 15: 169-

175.

Marthur A, Michalowicz B, Yang C, Aeppli D. Influence of periodontal bacteria and

disease status on Vß expression in T cells. J Periodontal Res 1995; 30: 369-373.

Marthur A, Michalowicz BS. Cell-mediated immune system regulation in periodontal

diseases. Crit Rev Oral Biol Med 1997; 8: 76-89.

Matthews JB, Mason GI. Immunoglobulin producing cells in human periapical

granulomas. Br J Oral Surg 1983; 21: 192-197.

Mathews RC, Burnie JP, Tabaqchali S. Immunoblot analysis of the serological

responses in invasive aspergillosis. J Clin Pathol 1984; 38: 1300-1303.

Mathews RC, Burnie JP, Tabaqchali S. Isolation of immunodominant antigens from

sera of patients with systemic candidosis and characterisation of serological response

to Candida albicans. J Clin Microbiol 1987; 25: 230-237.

Mathews R, Wells C, Burnie JP. Characterisation and cellular localisation of the

immunodominant 47-kDa antigen of Candida albicans. J Med Microbiol 1988; 27:

227-232.

327

Matsuki Y, yamamoto T, Hara K, Detection of inflammatory cytokine messenger

RNA (mRNA)-expressing cells in human inflamed gingiva by combined in situ

hybridization and immunohistochemistry. Immunology 1992; 76: 42-47.

Matsuki Y, Yamamoto T, Hara K. Localistation of interleukin-1 (IL-1) mRNA

expressing macrophages in human inflamed gingiva and IL-1 activity in gingival

crevicular fluid. J Periodontal Res 1993; 28: 35-42.

Mayer EP, Chen W-Y, Dray S, Teodorescu M. The identification of six mouse

lymphocyte subpopulations by their natural binding of bacteria. J Immunol 1978; 120:

167-173.

McArthur WP, Tsai CC, Baehni PC, Genco RJ, Taichman NS. Leukotoxic effects of

Actinobacillus actinomycetemcomitans modulation by serum conponents. J


McGhee JR, Kiyono H. In: Paul W, editor. Fundamental Immunology. Fourth edition.

Philadelphia: Lippincott-Raven; 1999.

McHeyzer-Williams MG, Ahmed R. B cell memory and the long-lived plasma cell.

Curr Opin Immunol 1999; 11: 172-179.

McMurray RW. Adhesion molecules in autoimmune disease. Semin Arthritis Rheum

1996; 25: 215-233.

Mealey BL. Periodontal implications: medically compromised patients. Ann

Periodontol 1996; 1: 256-321.

Mehra V, Bloom BR, Bajardi AC, et al. A major T cell antigen of Mycobacterium

leprae is a lOkDa heat shock cognate protein. J Exp Med 1992; 175: 275-284.

328

Meng HX, Zheng LF. T cells and T-cell subsets in periodontal diseases. J Periodontal

Res 1989; 24: 121-126.

Merville P, Pouteil NC, Wijdenes J, Potaux L, Touraine, Bancheraeau J. Detection of

single cells secreting IFN-gamma, IL-6 and IL-10 in irreversibly rejected human

kidney allografts, and their modulation of IL-2 and IL-4. Transplantation 1993; 55:

639-646.

Metchnikoff E. Immunity in the infectious diseases. First edition. New York:

Macmillan Press, 1905.

Michalowicz BS. Genetic and heritable risk factors in periodontal disease. J

Periodontol 1994; 65: 479-488.

Miller G, Lipman M. Release of infectious Epstein-Barr virus by transformed

marmoset leukocytes. Proc Natl Acad Sci USA 1973; 70: 190-194.

Miller DR, Lamster IB, Chasens Al. Role of the polymorphonuclear leukocyte in

periodontal health and disease. J Clin Periodontol 1984; 11: 1-15.

Miossec P, Cavender D, Ziff M. Production of interleukin-1 by human endothelial

cells. J Immunol 1986; 136: 2486-2491.

Mitchel GF. Co-evolution of parasites and adaptive immune responses. Immunol

Today 1991; 22: A2-A5.

Miyazaki A, Kobayashi T, Suzuki T, Yoshie H, Hara K. Loss of Fcy receptor and

impaired phagocytosis of polymorphonuclear leukocytes in gingival crevicular fluid. J


Modlin RL, Nutman TB. Type 2 cytokines and negative immune regulation in human

infections. Curr Opin Immunol 1993: 5: 511-517.

329

Mombelli A, McNabb H, Lang NP. Black-pigmenting Gram-negative bacteria in

periodontal disease. I. Topographic distribution in the human dentition. J Periodontal

Res 1991; 26: 301-307.

Moncla BJ, Motley ST, Braham P, Ewing L, Adams TH, Vermeulen MJ. Use of

synthetic oligonucleotide DNA probes for identification and direct detection of

Bacteroidesforsythus in plaque samples. J Clin Microbiol 1991; 29: 2158-2162.

Mooney J, Kinane DF. Comparison of humoral immune response to Porphyroniornas

gingivalis and Actinobacillus actinomycetemcomitans in adult periodontitis and

rapidly progressive periodontitis. Oral Microbiol Immunol 1994; 9: 321-326.

Mooney J, Adonogianaki E, Riggio MP, Takahashi K, Haerian A, Kinane DF. Initial

serum antibody titre to Porphyromonas gingivalis influences development of antibdy

avidity and success of therapy for chronic periodontitis. Infect Immun 1995; 63: 3411-

3416.

Moore WE, Holdeman LV, Smibert RM, Hash DE, Burmeister JA, Ranney RR.

Bacteriology of severe periodontitis in young adult humans. Infect Immun 1982; 38:

1137-1148.

Moore WEC, Holdeman LV, Cato EP, Smibert RM, Burmeister JA, Ranney RR.

Bacteriology of moderate (chronic) periodontitis in mature adult humans. Infect

Immun 1983; 42: 510-515.

Moore WE, Holdeman LV, Smibert RM, et al. Bacteriology of experimental gingivitis

in children. Infect Immun 1984; 46: 1-6.

Moore WEC, Holdeman LV, Cato EP, et al. Comparative bacteriology of juvenile

periodontitis. Infect Immun 1985; 48: 507-519.

330

Moore LV, Moore WE, Cato EP, et al. Bacteriology of human gingivitis. J Dent Res

1987; 66: 989-995.

Moore WE, Moore LH, Ranney RR, et al. The microflora of periodontal sites showing

active destructive progression. J Clin Periodontol 1991; 18: 729-739.

Moore WEC, Moore LVH. The bacteria of periodontal diseases, Periodontol 2000

1994; 5: 66-77.

Moritz A, Capelli D, Lantz MS, Holt SC, Ebersole JL. Immunization with

Porphyromonas gingivalis cysteine proteases: Effects on experimental gingivitis and

ligature-induced periodontitis in Macaw fascicularis. J Periodontol 1998; 69: 686-

697.

Morrison EC, Ramjford SP, Hill RW. Short-term effects of initial nonsurgical

periodontal treatment (hygienic phase). J Clin Periodontol 1980; 7: 199-211.

Moscow BS, Poison AM. Histologic studies on the extension of the inflammatory

infiltrate in human periodontitis. J Clin Periodontol 1991; 18: 534-542.

Mosman TR, Coffman RL. Heterogeneity of cytokine secretion patterns and functions

of helper T cells. Adv Immunol 1989; 46: 111-147.

Moss M, Beck J, Kaplan B. Exploratory case-control analysis of psychosocial factors

and adult periodontitis. J Periodontol 1996; 67 (suppl): 1060-1069.

Moter A, Hoenig C, Choi BK, Riep B, Gobel UB. Molecular epidemiology of oral

treponemes associated with periodontal disease. J Clin Microbiol 1998; 36: 1399-

1403.

Moudgil KD, Wang J, Yeung VP, Sercarz EE. Heterogeneity of the T cell response to

immunodominant determinants within hen eggwhite lysozyme of individual syngeneic

331

hybrid F1 mice: implications for autoimmunity and infection. J Immunol 1998; 161:

6046-6053.

Moughal NA. Adhesion molecule expression and cellular infiltrate within gingival

tissue. Thesis submitted to Faculty of medicine, University of Glasgow 1992.

Mouton C, Hammond PG, Slots J, Genco RJ. Serum antibodies to oral Bacteroides

asaccharolyticus (Bacteroides gingivalis): Relationship to age and periodontal

disease. Infect Immun 1981; 31: 182-192.

Mouton C, Desclauriers M, Allard H, Bouchard M. Serum antibodies to Bacteroicles

gingivalis in periodontitis: A longitudinal study. J Periodontal Res 1987; 22: 426-430.

Mühlemann HR, Son S. Gingival sulcus bleeding -a leading symptom in initial

gingivitis. Hely Odontol Acta 1971; 15: 107-113.

Munro JM, Pober JS, Cotran RS. Recruitment of neutrophils in the local endotoxin

response: Association with de novo endothelial expression of endothelial leukocyte

adhesion molecule-1. Lab Invest 1991; 64: 295-299.

Murphy JR. Diarrhoeal disease: current concepts and future challenges. Rapid

detection of enteric infections by measurement of enteric pathogen-specific antibody

secreting cells in peripheral blood. Trans R Soc Trop Med Hyg 1993; 3: 27-30.

Murphy P. The Neutrophil. New York: Plenum Publishing Corporation, 1976.

Murphy TF, Kyungcheol YI. Mechanisms of recurrent Otitis Media: Importance of the

immune response to bacterial surface antigens. Ann NY Acad Sci 1997; 830: 353-

360.

332

Murray PA, Burstein DA, Winkler JR. Antibodies to Bacteroides gingivalis in

patients with treated and untreated periodontal disease. J Periodontol 1989; 60: 96-

103.

Myers CD. Role of B cell antigen processing and presentation in the humoral immune

response. FASEB J 1991; 5: 2547-2553.

Mylotte JM, Stack RR, Murphy TR, Asirwatham J, Apicelle MA. Functional and

ultrastructural effects effects of nontypable Haemophilus influenzae in a hamster

trachea organ culture system. In vitro 1985; 21: 575-582.

Nagai A, Takahashi K, Sato N, Matsuo Y, Minowada J, Kurihara H et al. Abnormal

proportion of gamma delta T cells in peripheral blood is frequently detected in

patients with periodontal disease. J Periodontol 1993; 63: 963-967.

Naito Y, Okuda K, Takazoe I. Immunoglobulin G response to subgingival granm-

negative bacteria in human subjects. Infect Immun 1984; 45: 47-51.

Naito Y, Okuda K, Takazoe I. Detection of specific antibody in adult human

periodontitis sera to surface antigens of Bacteroides gingivalis. Infect Immun 1987;

55: 832-834.

Naito Y, Gibbons RJ. Attachment of Bacteroides gingivalis to Collagenous Substrata.

J Dent Res 1988; 67: 1075-1080.

Nakagawa T, Nakagawa S, Ishihara K, Yamada S, Machida Y, Okuda K Reactive

Antibodies in Sera from Pubertal and Adult Gingivitis Patients Against Various

Porphyromonas gingivalis Antigens. J Periodontal Res 1995; 30: 396-403.

Nakajima T, Yamazaki K, Hara K. Biased T cell receptor V gene usage in tissues with

periodontal disease. J Periodontal Res 1996; 31: 2-10.

333

Nara PL, Garrity R. Deceptive imprinting :a cosmopolitan strategy for complicating

vaccination. Vaccine 1998; 1: 1780-1787.

Narayanan AS, Page RC. Biochemical characterization of collagens synthesized by

fibroblasts derived from normal and diseased human gingiva. J Biol Chem 1976; 251:

5464-5471.

Nesheim S, Lee F, Sawyer M, et al. Diagnosis of human immunodeficiency virus infection by enzyme-linked immunospot assays in a prospectively followed cohort of infants of human immunodeficiency virus-seropositive women. Pediatr Infect Dis J

1992; 11: 635-639.

Ness KH, Morton TH, Dale BA. Identification of merkel cells in oral epithelium using

antikeratin and antineuroendocrine monoclonal antibodies. J Dent Res 1987; 66:

1154-1158.

Newman MG, Grinenco V, Weiner M, Angel I, Karge H, Nisengard R. Predominant

microbiota associated with periodontal health in the aged. J Periodontol 1978; 49:

553-559.

Niedbala WG, Stott DI. A comparison of three methods for production of human

hybridomas secreting autoantibodies. Hybridoma 1998; 17: 299-304.

Niekrash CE, Patters MR. Simultaneous assessment of complement components C3,

C4 and B and their cleavage products in human gingival fluid. II. Longitudinal

changes during periodontal therapy. J Periodontal Res 1985; 20: 268-275.

Nieminen T, Virolainen A, Käyhty H, et al. Antibody secreting cells and their relation

to humoral antbodies in serum and in nasopharyngeal aspirates in children with

pneumococcal acute otitis media. J Infect Dis 1996; 173: 136-141.

334

Nisengard RJ. The role of immunology in periodontal disease. J Periodontol 1977; 48:

505-516.

Nisengard RJ, Newman MN, Myers D, Horikoshi A. Humoral immunologic responses in idiopathic juvenile periodontitis (periodontosis). J Periodontol 1980; 51: 30-33.

Nisengard RJ, Newman MG. Oral Microbiology and Immunology. Second Ed. WB

Saunders company. 1994.

Nishikata M, Yoshimura F. Characterisation of Porphyromonas gingivalis

Haemagluttinin as a Protease. Biochim Biophys Res Commun 1991; 178: 336-342.

Noelle RJ, Snow EC. Cognate interactions between helper T cells and B cells.

Immunol Today 1990; 11: 361-368.

Offenbacher S, Heasman P, Collins J. Modulation of host PGE2 secretion as a

determinant of periodontal disease. J Periodontol 1993; 64: 432-444.

Offenbacher S, Katz V, Fertik G, et al. Periodontal infection as a possible risk factor

for preterm low birth weight. J Periodontol 1996; 67: 1103-1113.

Offenbacher S. Periodontal Diseases: Pathogenesis. Ann Periodontol 1996; 1: 821-

878.

Ogawa T, McGhee ML, Moldoveanu Z, et al. Bacteroides-specific IgG and IgA

subclass antibody-secreting cells isolated from chronically inflamed gingival tissues.

Clin Exp Immunol 1989a; 76: 103-110.

Ogawa T, Tarkowksi A, McGhee ML et al. Analysis of human IgG and IgA subclass

antibody-secreting cells from localized chronic inflammatory tissue. J Immunol

1989b; 142: 1150-1158.

335

Ogawa T, Kono Y, McGhee ML, et al. Porphyromonas gingivalis-specific serum IgG

and IgA antibodies originate from immunoglobulin-secreting cells in inflamed

gingiva. Clin Exp Immunol 1991; 83: 237-244.

Ohta K, Makinen KK, Loesche WJ. Purification and characterisation of an enzyme

produced by Treponema denticola capable of hydrolysing synthetic trypsin substrates.

Infect Immun 1986; 53: 213-220.

Okada Y, Suzuki T, Hosaka Y. Interaction between influenza virus and Ehrlich's

tumor cells. III. Fusion phenomenon of Ehlich's tumor cells by the action of HVJ Z

strain. Med J Osaka Univ 1957; 7: 709.

Okada H, Kida T, Yamagami H. Characterisation of the immunocompetent cells in

human advanced periodontitis. J Periodontal Res 1982; 17: 472-473.

Okada H, Kida T, Yamagami H. Identification and distribution of immunocompetent

cells in inflamed gingiva of human chronic periodontitis. Infect Immun 1983; 41: 365.

Okada H, Murakami S. Cytokine expression in periodontal health and disease. Crit

Rev Oral Biol Med 1998; 9: 248-266.

Okuda K, Kato T, Naito Y, Takazoe I. Precipitating antibody against

lipopolysaccharide of Haemophilus actinomycetemcomitans in human serum. J Clin

Microbiol 1986; 24: 846-848.

Okuda K, Takazoe I. The role of Bacteroides gingivalis in periodontal disease. Adv

Dent Res 1988; 2: 260-268.

Oliver RC, Holm-Pedersen P, Löe H. The correlation between clinical scoring,

exudate measurements and microscopic evaluation of inflammation in the gingiva. J

Periodontol 1969; 40: 201-209.

336

Olsen F. Attachment of Treponema denticola to cultured human epithelial cells.

Scand J Dent Res 1984; 92: 55-63.

Olsen I, Shah HN, Gharbia SE. Taxonomy and biochemical characteristics of

Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis. Periodontol

2000 1999; 20: 14-52.

Olsson I, Kaplan HS. Human-human hybridomas producing monoclonal antibodies of

predefined antigenic specificity. Proc Natl Acad Sei USA 1980; 77: 5429-5433.

Omar AA, Newman HN, Bulman J, Osborn J. Associations between subgingival

plaque bacteria; morphotypes and clinical indices? J Clin Periodontol 1991; 18: 555-

566.

Orr N, Robin G, Lowell G, Cohen D. Presence of specific immunoglobulin A-

secreting cells in peripheral blood after natural infection with Shigella sonnei. J Clin

Microbiol 1992; 30: 2165-2168.

Osborn L, Hession C, Tizard R, et al. Direct cloning of vascular cell adhesion

molecule 1, a cytokine-induced endothelial protein that binds to lymphocytes. Cell

1989; 59: 1203-1211.

Page MI, King EO. Infection due to Actinobacillus actnomycetemcomitans and

Haemophilus aphrophilus. N Engl J Med 1966; 275: 181-188.

Page RC, Ammons WF. Collagen turnover in the gingiva and other mature connective

tissue of the marmoset sanguis oedipus. Arch Oral Biol 1974; 19: 651-658.

Page RC, Simpson DM, Ammons WF. Host tissue response in chronic inflammatory

periodontal disease. IV. The periodontal and dental status of a group of aged great

apes. J Periodontol 1975; 46: 144-155.

337

Page RC, Schroeder HE. Pathogenesis of inflammatory periodontal disease. A

summary of current work. Lab Invest 1976; 33: 235-249.

Page RC. Oral health status in the united states: Prevalence of inflammatory

periodontal diseases. J Dent Educ 1985; 49: 354-367.

Page RC. Gingivitis. J Clin Periodontol 1986; 13: 345-355.

Page RC. The role of inflammatory mediators in the pathogenesis of periodontal

disease. J Periodontal Res 1991; 26: 230-242.

Page RC, Rose M, Yacoub M, Pigott R. Antigenic heterogeneity of vascular

endothelium. Am J Pathol 1992a; 141: 673-683.

Page RC, Sims TJ, Moncla BJ, Bainbridge RP, Engel LD. Reactive antigens of

periodontopathic bacterium A. actinomycetemcomitans. Adv Exp Med Biol 1992b;

327: 243-253.

Page RC, Beck JD. Risk assessment for periodontal diseases. Int Dent J 1997; 47: 61-

87.

Pamer EB, Sijts AJ, Villaneuva MS, Busch DH, Vijh S. MHC class I antigen

processing of Listeria monocytogenes proteins: implications for dominant and

subdominant CTL responses. Immunol Rev 1997; 158: 129-136.

Panchanathan V, Naidu BR, Devi S, Di Pasquale A, Mason T, Pang T. Immunogenic

epitopes of Salmonella typhi GroEL heat shock protein rective with both monoclonal

antibody and patients sera. Immun Lett 1998; 62: 105-109.

Papapanou PN, Wennstrom JL, Grondhal K. Periodontal status in relation to age and

tooth type. A cross sectional radiographic study. J Clin Periodontol 1988; 15: 469-

478.

338

Papapanou PN. Periodontal diseases: Epidemiology. Ann Periodontol 1996; 1: 1-36.

Pasanen VJ, Mäkelä 0. Effect of the number of haptens coupled to each erythrocyte

on haemolytic plaque formation. Immunology 1969; 16: 399-407.

Pattres MR, Niekrash CE, Lang NP. Assessment of complement cleavage in gingival

fluid during experimental gingivitis. J Clin Periodontol 1989; 16: 33-37.

Pavloff N, Potempa J, Pike RN, et al. Molecular cloning and structural

characterization of the Arg-gingipain proteinase of Porphyromonas gingivalis.

Biosynthesis as a proteinase-adhesin polyprotein. J Biol Chem 1995; 270: 1007-1010.

Payne WA, Page RC, Ogilivie AL, Hall WB. Histopathologic features of the initial

and early stages of experimental gingivitis in man. J Periodontal Res 1975; 10: 51-64.

Peitz GA. Trounstine ML. Moore KW. Cloned and expressed human Fc receptor for

IgG mediates anti-CD3-dependent lymphoproliferation. J Immunol 1988; 141: 1891-

1896.

Peng TK. Serum antibodies to native and denatured type I and II collagen in patients

with periodontal disease. Proc Natl Sei Counc B Repub China 1988; 12: 21-26.

Pereira CM, Yamaauchi LM, Levin MJ, da Silveira JF, Castilho BA. Mapping of B

cell epitopes in an immunodominant antigen of Trypanosoma cruzi using fusions to

the Escherichia coli LamB protein. FEMS Microbiol Lett 1998; 164: 125-131.

Perez-Perez GI, Thiberge JM, Labigne A, Blaser MJ. Relationship of immune

response to heat shock protein A and characteristics of Heliobacter pylori-infected

patients. J Infect Dis 1996; 174: 1046-1050.

339

Petit MD, Stashenko P. Extracts of periodontopathic microorganisms lack functional

superantigenic activity for murine T cells. J Periodontal Res 1996; 31: 517-524.

Pichichero ME, Anderson P, Loeb M, Smith DH. Do pili play a role in pathogenicity

of Haemophilus influenzae type b? Lancet 1982; ii: 960-962.

Pike R, McGraw W, Potempa J, Travis J. Lysine and Arginine specific proteinases

from Porphyromonas gingivalis. J Biol Chem 1996; 269: 406-411.

Pinard A. Gingivitis in preganacy. Dent Register 1877; 31: 258-259.

Pitzalis C, Kingsley C, Haskard D, Panayi G. The preferential accumulation of helper-

induced T lymphocytes in inflammatory lesions: evidence for regulation by selective

endothelial and homotypic adhesion. Eur J Immunol 1988; 18: 1397-1404.

Platt D, Crosby RG, Dalbow MH. Evidence for the presence of immunoglobulins and

antibodies in inflamed periodontal tissues. J Periodontol 1970; 41: 215-222.

Poison AM, Goodson JM. Periodontal diagnosis. Current status and future needs. J

Periodontol 1985; 56: 25-34.

Potempa J, Travis J. Porphyromonas gingivalis proteinases in periodontitis, a review.

Acta Biochim Pol 1996; 43: 455-465.

Prabhu A, Michalowicz BS, Marthur A. Detection of local and systemic cytokines in

adult periodontitis. J Periodontol 1996; 67: 515-522.

Preber H, Bergstrom J. Effect of cigarette smoking on periodontal healing following

surgical therapy. J Clin Periodontol 1990; 17: 324-328.

Preber H, Berström J, Linder LE. Occurrence of periodpathogens in smoker and non-

smoker patients. J Clin Periodontol 1992; 19: 667-671.

340

Preston A, Mandrell RE, Gibson BW, Apicella MA. The Lipooligosaccharides of

Pathogenic Gram-negative Bacteria. Crit Rev Microbiol 1996; 22: 139-180.

Preus HR, Anerud A, Boysen H, Dunford RG, Zambon JJ, Löe H. The natural history

of periodontal disease. The correlation of selected microbiological parameters with

disease severity in Sri Lankan tea workers. J Clin Periodontol 1995; 22: 674-678.

Pridmore RD. New and versatile cloning vectors with kanamycin-resistance marker.

Gene 1987; 56: 309-312.

Pulver WH, Taubman MA, Smith DJ. Immune components in human dental

periapical lesions. Arch Oral Biol 1978; 23: 435-443.

R&D Systems Inc. Cell Adhesion Molecules (Poster) 1994.

Ranney RR. Immunologic mechanisms of pathogenesis in periodontal diseases: as

assessment. J Periodontal Res 1991; 26: 243-254.

Rateitschak-Plüss EM, Hefti A, Lörtscher R, Thiel G. Initial observation that

cyclosporin-A induces gingival enlargement in man. J Clin Periodontol 1983; 10: 237-

246.

Reinhardt RA, McDonald TL, Bolton RW, Dubois LM, Kaldahl WB. IgG subclasses

in gingival crevicular fluid from active versus stable periodontal sites. J Periodontol

1989; 60: 44-50.

Reinhardt RA, Masada MP, Kaldahl WB et al. Gingival fluid IL-I and IL-6 levels in

refractory periodontitis. J Clin Peridontol 1993; 20: 225-231.

Reinholdt J, Bay I, Svejgaard A. Association between HLA-antigens and periodontal

disease. Part I. J Dent Res 1977; 56: 1261-1263.

341

Reme T, Portier M, Frayssinouf F, et al. T cell receptor expression and activation of

synovial lymphocyte subsets in patients with rheumatoid arthritis. Phenotyping of

multiple synovial sites. Arthritis Rheum 1990; 33: 485-492.

Reth M. Antigen receptors on B lymphocytes. Annu Rev Immunol 1992; 10: 97-121.

Richards D, Rutherford RB. The effects of interleukin-1 on collagenolytic activity and

prostaglandin-E secretion by human peri odontal- ligament and gingival fibroblasts.

Arch Oral Biol 1988; 33: 237-243.

Rietschel ET, Wollenweber HW, Brade H, et al. Chemistry of Endotoxin: Structure

and Conformation of the Lipid A Component of Lipopolysaccharides. In: Rietschel

ET, editor. Handbook of Endotoxin, vol 1. Elsevier, 1984.

Riviere GR, Elliot KS, Adams DF. Relative proportions of pathogen-related oral

spirochetes (PROS) and Treponema denticola in supragingival and subgingival plaque

from patients with periodontitis. J Periodontol 1992; 63: 131-136.

Riviere GR, Smith KS, Carranza JrN, Tzagaroulaki E, Kay SL, Dock M. Subgingival

distribution of Treponema denticola, Treponema socranskii and pathogen-related oral

spirochetes: Prevalence and relationship to periodontal sampled sites. J Periodontol

1995; 66: 829-837.

Riviere GR, Smith KS, Tzagaroulaki E, et al. Periodontal status and detection

frequency of bacteria at sites of periodontal health and gingivitis. J Periodontol 1996;

67: 109-115.

Robinson A, Duggleby CJ, Gorringe AR, Livey I. Antigenic variation in Bordetella

pertussis. In: Birbeck TH, Penn CW, editors. Antigenic variation in Infectious

Disease. Washington D. C: IRC Press, 1986: 147-161.

342

Rosales C, Brown EJ. Neutrophil receptors and the modulation of the immune

response. In: Abramson JS, Wheeler JG, editors. The Neutrophil. New York: Oxford

University Press, 1993.

Rosan B, Slots J, Lamont RJ, Listgarten MS, Nelson GM. Actinobacillus

actinomycetemcomitans fimbriae. Oral Microbiol Immunol 1988; 3: 58-63.

Rosen A, Bergeley P, Jondal M et al. Polyclonal Ig production after Epstein-Barr

virus infection of human lymphocytes in vitro. Nature 1977; 267: 52-54.

Rossomando EF, Kennedy JE, Hadjmicheal J. Tumour necrosis factor alpha in

gingival crevicular fluid as a possible indicator of periodontal disease in humans. Arch

Oral Biol 1990; 35: 431-434.

Roulsten D, Schwartz S, Cohen MM, Suzuki JB, Weitkamp LR, Boughman JA.

Linkage analysis of dentinogenesis imperfecta and juvenile periodontitis creating a5

point map of 4q. Am J Hum Genet 1985; 37: A206 (abstr).

Rüdin HJ, Overdiek HF, Rateitschak K-H. Correlation between sulcus fluid rate and

clinical and histological inflammation of the marginal gingiva. Hely Odontol Acta

1970; 14: 21-26.

Saglie FR, Marfany A, Carnargo P. Intragingival occurrence of Acinobacillus

actinomycetemcomitans and Bacteroides gingivalis in active destructive periodontal

lesions. J Periodontol 1988; 59: 259-265.

Salvi GE, Lawrence HP, Offenbacher S, Beck JD. Influence of risk factors on the

pathogenesis of periodontitis. Periodontol 2000 1997; 14: 173-201.

Salvi GE, Beck JD, Offenbacher S. PGE2, IL-1 beta and TNF-alpha responses in

diabetics as modifiers of periodontal disease expression. Ann Periodontol 1998; 3: 40-

50.

343

Sanchez L, Calvo M, Brock J. Biological role of lactoferrin. Arch Dis Child 1992; 67:

657-661.

Sandholm L, Tolo K, Olsen I. Salivary IgG, a parameter of periodontal disease

activity? High responders to Actinobacillus actinomycetemcomitans Y4 in juvenile

and adult periodontitis. J Clin Periodontol 1986; 14: 289-294.

Sandholm L, Tolo K. Serum antibody levels to 4 periodontal pathogens remain

unaltered after mechanical therapy of juvenile periodontitis. J Clin Periodontol 1987;

13: 646-650.

Sandros J, Madianos PN, Papapanou PN. Cellular events concurrent with

Porphyromonas gingivalis invasion of oral epithelium in vitro. Eur J Oral Sci 1996;

104: 363-371.

Saunders RL de CH, Röckert HÖE. vascular supply of dental tissues including

lymphatics. In: Miles AEW, editor. Structural and chemical organization of teeth.

Volume 1. London: Academic Press, 1967: 199-245.

Savitt ED. Socransky SS. Distribution of certain subgingival microbial species in

selected periodontal conditions. J Periodontal Res 1984; 19: 111-123.

Saxen I, Koskimies S. Juvenile periodontitis - no linkage with HLA antigens. J


Scannipieco FA, Kornman KS, Coykendall AL. Observation of fimbriae and flagella

in dispersed subgingival dental plaque and fresh bacterial isolates from periodontal

diseases. J Periodontal Res 1983; 18: 620-633

Sannapieco FA, Berey EJ, Reddy MS, Levine MJ. Characterization of salivary a-

amylase binding to Streptococcus sanguis. Infect Immun 1989; 57: 2853-2863.

344

Scannapieco FA. Saliva-bacterium interactions in oral microbial ecology. Crit Rev

Oral Biol Med 1994; 5: 203-248.

Scannapieco FA, Torres GI, Levine MJ. Salivary amylase promotes adhesion of oral

streptococci to hydroxyapatite. J Dent Res 1995; 74: 1360-1366.

Schenk K. IgG, IgA and IgM serum antibodies against lipopolysaccharide from

Bacteroides gingivalis in periodontal health and disease. J Periodontal Res 1985; 20:

368-377.

Schenk K, Michaelsen TE. IgG subclass distribution of serum antibodies against

lipopolysaccharide from Bacteroides gingivalis in periodontal health and disease.

APMIS 1987; 95: 41-46.

Schenkein HA, Cianciola L, Genco R. Complement activation in gingival pocket fluid

from patients with periodontosis and severe periodontitis. IADR abstract 1976.

Schenkein HA, Genco RJ. Gingival fluid and serum in periodontal diseases. II.

Evidence for activation of complement components C3, C3 proactivator and C4 in

gingival fluid. J Periodontol 1977; 48: 778-784.

Schenkein HA. The effect of periodontal proteolytic Bacteroides species on proteins

of the human complement system. J Periodontal Res 1988; 23: 187-192.

Schenkein HA, Fletcher HM, Bodnar M, Macrina FL. Increased opsonisation of a

prtH defective mutant of Porphyromonas gingivalis W83 is caused by reduced

degradation of complement derived opsonins. J Immunol 1995; 154: 5331-5337.

Schifferle RE, Wilson ME, Levine MJ, Genco RJ. Activation of serum complement

by polysaccharide-containing antigens of Porphyromonas gingivalis. J Periodontal

Res 1993; 28: 248-254.

345

Schneider TF, Toto PD, Gargiulo AW, Pollock RJ. Specific bacterial antibodies in the

inflamed human gingiva. Periodontics 1966; 4: 53-57.

Schonfeld SE, Kagan JM. Determination of the specificity of plasma cells for

bacterial antigens in situ. Infect Immun 1980; 27: 947-952.

Schonfeld SE, Kagan JM. Specificity of gingival plasma cells for bacterial somatic

antigens. J Periodontal Res 1982; 17: 60-69.

Schroeder HE. Melanin containing organelles in cells of the human gingiva. J

Periodontal Res 1969a; 4: 1-18.

Schroeder HE. Melanin containing organelles in cells of the human gingiva. II

Keratinocytes. J Periodontal Res 1969b; 4: 235-247.

Schroeder H, Graf-de-Beer M, Attström R. Initial gingivitis in dogs. J Periodontal Res

1975; 110: 128.

Schroeder HE. The periodontium. Handbook of microscopic anatomy. Volume V/5.

Berlin: Springer, 1986.

Schroeder HE, Listgarten MA. The gingival tissues: the architecture of periodontal

protection. Periodont 2000 1997; 13: 91-120.

Schryvers AB, Morris U. Identification and characterisation of the human lactoferrin

binding protein of Neisseria meningitidis. Infect Immun 1988; 56: 1144-1149.

Schwaber JF, Posner MR, Schlossman SF, Lazarus H. Human-human hybrids

secreting pneumococcal antibodies. Hum Immunol 1984; 9: 137-143.

346

Sedgwick JD, Holt PG. A solid-phase immunoenzymatic technique for the

enumeration of specific antibody-secreting cells. J Immunol Methods 1983; 48: 301-

309.

Sercarz EE. The architectonics of immune dominance: The aleatory effects of

molecular position on the choice of antigenic determinants. Chem Immunol 1989;

46: 169-185.

Seifert M, Schoenherr G, Roggenbuck D, Marx U, von Baehr R. Generation and

characterisation of a human monoclonal IgM antibody that recognises a conserved

epitope shared by lipopolysaccharides of different Gram negative bacteria. Hybridoma

1996; 15: 191-198.

Seymour G, Greenspan J. The phenotypic characterization of lymphocyte

subpopulations in established human periodontal disease. J Periodontal Res 1979; 14:

39-44.

Seymour GJ, Powell RN, Davies WIR. Conversion of a stable T-cell lesion to a

progressive B-cell lesion in the pathogenesis of chronic inflammatory periodontal

disease: a hypothesis. J Clin Periodontol 1979; 6: 267-277.

Seymour GJ, Crouch MS, Powell RN et al. The identification of lymphoid cell

subpopulations in sections of human lymphoid tissue in gingivitis in children using

monoclonal antibodies. J Periodontal Res 1982; 17: 247-256.

Seymour GJ, Powell RN, Cole KL et al. Experimental gingivitis in humans. A

histochemical and immunological characterisation of the lymphoid cell

subpopulations. J Periodontal Res 1983; 18: 375-385.

Seymour RA. Calcium channel blockers and gingival overgrowth. Br Dent J 1991;

170: 376-379.

347

Seymour RA, Jacobs DJ. Cyclosporin and the gingival tissues. J Clin Periodontol

1992; 19: 1-11.

Seymour GJ, Gemmel E, Reinhardt RA, Eastcott J, Taubman MA.

Immunopathogenesis of chronic inflammatory periodontal disease: cellular and

molecular mechanisms. J Periodontal Res 1993; 28: 478-486.

Seymour RA, Thomason JM, Ellis JS. The pathogenesis of drug-induced gingival

overgrowth. J Clin Periodontol 1996; 23: 165-175.

Shalaby MR, Waage A, Espevik T. Cytokine regulation of interleukin 6 production by

human endothelial cells. Cell Immunol 1989; 121: 372-382.

Shapira L, Eizenberg S, Sela MN, Soskolne A, Brautbar H. HLA-A9 and B-15 are

associated with the generalized form but not localized form of early onset periodontal

disease. J Periodontol 1994; 65: 219-223.

Shah HN, Gharbia SE. Biochemical and chemical studies on strains designated

Prevotella intermedia and proposal of a new pigmented species, Prevotella nigrescens

sp. nov. Int J Syst Bacteriol 1992; 42: 542-546.

Shenk K, Poppelsdorf D, Denis C, Tollefsen T. Levels of salivary IgA antibodies

reactive with bacteria from dental plaque are associated with susceptibility to

experimental gingivitis. J Clin Periodontol 1993; 20: 411-417.

Shenkels LC, Veerman EC, Nieuw Amerongen AV. Biochemical composition of

human saliiva in relation to other mucosal fluids. Crit Rev Oral Biol Med 1995; 6:

161-175.

Shenker BJ, Wannberg S, Vitale L, Kieba I, Lally ET. Reconstruction of severe

combined immunodeficient mice with lymphocytes from patients with localised

juvenile periodontitis. J Periodontal Res 1993; 28: 487-490.

348

Shields JM, Haston WS. Behaviour of neutrophil leucocytes in uniform

concentrations of chemotactic factors: contraction waves, cell polarity and persistence.

J Cell Sci 1985; 74: 75-93.

Shimizu N, Ogura N, Yamaguchi M et al. Stimulation by interleukin-1 of interleukin-

6 production by human peridontal ligament cells. Arch Oral Biol 1992; 37: 743-748.

Shinnick TM, Vodkin MH, Williams JC. The Mycobacteriunz tuberculosis 65-

Kilodalton antigen is a heat shock protein which corresponds to common antigen and

to Escherichia coli GroEL protein. Infect Immun 1988; 56: 446-451.

Shinnick TM. Heat shock proteins as antigens of bacterial and parasitic pathogens.

Curr Top Microbiol Immunol 1991; 167: 145-160.

Shirahata S, Katakura Y, Teruya K. Cell hybridization, hybridomas and human

hybridomas. Methods Cell Biol 1998; 57: 111-145.

Simmons S, Makgoba MW, Seed B. ICAM, an adhesion ligand of LFA-1, is

homologous to the neural cell adhesion molecule NCAM. Nature 1988; 331: 624-627.

Simonson LG, Goodman CH, Bial JJ, Morton HE. Quantitative relationship of

Treponema denticola to severity of periodontal disease. Infect Immun 1988; 56: 726-

728.

Simonson LG, Robinson PJ, Pranger RJ, Cohen ME, Morton HE. Treponema

denticola and Porphyromonas gingivalis as prognostic markers following periodontal

treatment. J Periodontol 1992; 63: 270-273.

Sims TJ, Moncla BJ, Darveau RP, Page RC. Antigens of Actinobacillus

actinomycetemcomitans recognised by patients with juvenille periodontitis and

periodontally normal subjects. Infect Immun 1991; 59: 913-924.

349

Sims TJ, Mancl LA, Braham PH, Page RC. Antigenic variation in Bacteroides

forsythus detected by a checkerboard enzyme-linked immunosorbent assay. Clin

Diagn Lab Immunol 1998; 5: 725-731.

Sissons CH, Wong L, Cutress TW. Patterns and rates of growth of microcosm dental

plaque biofilms. Oral Microbiol Immunol 1995; 10: 160-167.

Sjöström K, Ou J, Whitney C, et al. Effect of treatment on titer, function and antigen

recognition of serum antibodies to Actinobacillus actinomycetemcomitans in patients

with rapidly progressive periodontitis. Infect Immun 1994; 62: 145-151.

Skerka C, Irving SG, Bialonski A, Zipfel PF. Cell type specific expression of

members of the IL-8/NAP-1 gene family. Cytokine 1993; 5: 122-116.

Skopek RJ, Liljemark WF. The influence of saliva on interbacterial adherence. Oral


Slaney JM, Rangarajan M, Aduse-Opoku J, et al. Recognition of the carbohydrate

modifications to the RgpA protease of Porphyromonas gingivalis by periodontal

patient serum IgG. J Periodontal Res 2000. In Press.

Slots J. Importance of black pigmented Bacteroides in human periodontal disease.

1982 In: Genco RJ, Mergenhagen SE, eds. Host-parasite interactions in periodontal

diseases. American Society for Microbiology, Washington. pp 27-45.

Slots J. Microflora in the healthy gingival sulcus in man. Scand J Dent Res 1977; 85:

247-254.

Slots J, Moenbo D, Langebaek J, frandsen A. Microbiota of gingivitis in man. Scand J

Dent Res 1978; 86: 174-181.

350

Slots J, Reynolds HS, Genco RJ. Actinobacillus actinomycetemcomitans in human

periodontal disease: a cross-sectional, microbiological investigation. Infect immun

1980; 29: 1013-1020.

Slots J, Genco RJ. Black pigmented bacteroides species, capnocytophaga species and

actinobacillus actinomycetemcomitans in human periodontal disease: Virulence

factors in colonisation, survival and tissue destruction. J Dent Res 1984; 63: 412-421.

Slots J, Bragd L, Wikström M, Dahlen G. The occurence of Actinobacillus

actinomycetemcomitans, Bacteroides gingivalis and Bacteroides intermedius in

destructive periodontal disease in adults. J Clin Periodontol 1986; 13: 570-577.

Slots J, Listgarten MA. Bacteroides gingivalis, Bacteroides intermedius and

Actinobacillus actinomycetemcomitans in human periodontal diseases. J Clin

Periodontol 1988; 15: 85-93.

Slots J. Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis in

periodontal disease: introduction. Periodontol 2000 1999; 20: 7-13.

Smith, CW, Hollers JC, Patrick RA, Hassett C. Motility and adhesiveness in human

neutrophils: Effects ofchemotactic factors. J Clin Invest 1979; 63: 221-229.

Smith DJ, Gadalla LM, Ebersole JL, Taubman MA. Gingival crevicular antibody to

oral microorganisms. III. Association of gingival homogenate and gingival crevicular

fluid antibody levels. J Periodontal Res 1985; 20: 357-367.

Smith G, Mathews JB, Smith AJ, Browne RM. Immunoglobulin-producing cells in

human odontogenic cysts. J Oral Pathol 1987; 16: 45-48.

Smith KG, Hewitson TD, Nossal GJ, Tarlinton DM. The phenotype and fate of the

antibody-forming cells of the splenic foci. Eur J Immunol; 1996; 26: 444-448.

351

Snow EC, Noelle RJ. The role of direct cellular communication during the

development of a humoral immune response. Adv Cancer Res 1993; 62: 241-266.

Socransky SS. Relationship of bacteria to the etiology of periodontal disease. J Dent

Res 1970; 49: 203-222.

Socransky SS. Microbiology of periodontal disease - Present status and future

considerations. J Periodontol 1977; 48: 497-504.

Socransky SS, Tanner ACR, Haffajee AD, Hillman JD, Goodson JM. Present status of

studies on the microbial etiology of periodontal diseases. In: Genco RJ, Mergenhagen,

editors. Host-parasite interactions in periodontal diseases. Washington DC: American

Society for Microbiology, 1982: 1-12.

Socransky SS, Haffajee AD, Dzink JL, Hillman JD. Associations between microbial

species in subgingival plaque samples. Oral Microbiol Immunol 1988; 3: 1-7.

Socransky SS, Haffajee AD. Microbiological risk factors for destructive periodontal

diseases. In: Bader JD, ed. Risk assessment in dentistry. Chapel Hill: University of

North Carolina Dental Ecology, 1990: 79-90.

Socransky SS, Haffajee AD. Microbial mechanisms in the pathogenesis of destructive

periodontal diseases: a critical assessment. J Periodontal Res 1991; 26: 195-212.

Socransky SS, Haffajee AD. The bacterial etiology of destructive periodontal disease:

Current concepts. J Periodontol 1992; 63: 332-331.

Socransky SS, Haffajee AD. Effect of therapy on periodontal infections. J Periodontol

1993; 64: 754-759.

Socransky SS, Haffajee AD. Evidence of bacterial etiology: a historical perspective.

Periodontol 2000 1994; 5: 7-25.

352

Sperber GH. Oral contraceptive hypertrophic gingivitis. J Dent Assoc S Africa 1969;

24: 37-40.

Stamm JW. Epidemiology of gingivitis. J Clin Periodontol 1986; 13: 360-366.

Stashenko P, Dewhirst FE, Peros WJ, Kent RL, Age JM. 1987. Synergistic

interactions between interleukin-1, tumor necrosis factor and lymphotoxin in bone

resorption. J Immunol 1987; 138: 1464-1468.

Stashenko P, Fujiyoshi P, Obernesser MS, Prostak L, Haffajee AD, Socransky SS.

Levels of interleukin-1 beta in tissue from sites of active periodontal disease. J Clin

periodontol 1991; 18: 548-554.

Stashenko P, Jandinski JJ, Fujiyoshi P, Rynar J, Socransky SS. Tissue levels of bone

resorptive cytokine in periodontal disease. J Periodontol 1991; 62: 504-509.

Staunton DE, Merluzzi VJ, Rothlein R, Barton R, Marlin SD, Springer TA. A cell

adhesion molecule, ICAM-1, is the major surface receptor for rhinoviruses. Cell 1989;

56: 849-853.

Staunton DE, Dustin ML, Springer TA. Functional cloning of ICAM-2, a cell

adhesion ligand for LFA-1 homologous to ICAM-1. Nature 1989; 339: 61-64.

Staunton DE, Dustin ML, Erickson HP, Springer TA. The arrangement of the

immunoglobulin like domains of ICAM-1 and the binding sites for LFA-1 and

rhinovirus. Cell 1990; 61: 243-254.

Steffen MJ, Holt SC, Ebersole JL. Porphyromonas gingivalis induction of mediator

and cytokine secretion by human gingival fibroblasts. Oral Microbiol Immunol 1999

(in press).

353

Stein SH, Hendrix CL. Interleukin-10 promotes anti-collagen antibody production in

gingival mononuclear cells. J Dent Res 1996; 75: 158 (abstract).

Stein SH, Hart TE, Hoffman WH, Hendrix CL, Gustke CJ, Watson SC. Interleukin-10

promotes anti-collagen antibody production in type I diabetic peripheral B

lymphocytes. J Periodontal Res 1997; 32: 189-195.

Steinitz M, Klein G, Koskimies S, Makela 0. EB virus-induced B lymphocyte cell

lines producing specific antibody. Nature 1977; 269: 420-422.

Stern IB. Electron microscopic observatons of oral epithelium. I. Basal cells and the

basement membrane. Periodontics 1965; 3: 224-238.

Stern MH, Dreizen S, Mackler BF, Levy BM. Antibody-producing cells in human

periapical granulomas and cysts. J Endodon 1981; 7: 447-452.

Stevens RH, Macy E, Morrow C, Saxon A. Characterization of a circulating

subpopulation of spontaneous antitetanus toxoid antibody producing B cells following

in vivo booster immunization. J Immunol 1979; 122; 2498-2504.

Stoolman LM. 1989. Adhesion molecules controlling lymphocyte migration. Cell

1989; 56: 907-910.

Stossel TP. Phagocytosis. N Engl J Med 1974; 290: 713-723,774-779.

Stoufi ED, Taubman MA, Ebersole JL, Smith DJ, Stashenko PP. Phenotypic analyses

of mononuclear cells recovered from healthy and diseased human periodontal tissues.

J Clin Immunol 1987; 7: 235-245.

Stoufi ED, Taubman MA, Ebersole JL, Smith DJ. A method for studying

immunoglobulin synthesis by gingival cells. Oral Microbiol Immunol 1988; 3: 109-

112.

354

Stricker EAM, Tiebout RF, Lelie PN, Zeijlemaker WP. A human monoclonal IgG

anti-hepatitis B surface antibody: Production, properties and applications. Scand J

Immunol 1985; 22: 337-343.

Stull TL, Mendelman PM, Haas JE, Schoenborn MA, Mack KD, Smith AL.

Characterisation of Haemophilus influenzae type b fimbriae. Infect Immun 1984; 46:

787-796.

Sugiyama T, Yabana T, Yachi A. Mucosal immune response to Heliobacter pylori

and cytotoxic mechanism. Scand J Gastroenterol suppl 1995; 208: 30-32.

Sundqvist G, Figdor D, Hanstrom L, Sorlin S, Sandstrom G. Phagocytosis and

virulence of different strains of Porphyromonas gingivalis. Scand J Dent Res 1991;

99: 117-129.

Suzuki JB, Martin SA, Vincent JW, Falkler WA Jr. Local and systemic production of

immunoglobulins to periodontopathogens in periodontal disease. J Periodontal Res

1984; 19: 599-603.

Syed SA, Loesche WJ. Bacteriology of human experimental gingivitis: effect of

plaque age. Infect Immun 1978; 21: 821-829.

Syrjanen S, Markkanen H, Syrjanen K. Inflammatory cells and their subsets in lesions

and juvenile periodontitis. A family study. Acta Odontol Scand 1984; 42: 285-292.

Takada H, Kimura S, Hamada S. Induction of inflammatory cytokines by a soluble

moiety prepared from an enzyme lysate of Actinomyces viscosus cell walls. J Med

Microbiol 1993; 38: 395-400.

Takahashi K, Takigawa M, Takashiba S, Nagai A, Miyamoto M, Kurihara H,

Murayama Y. Role of cytokine in the induction of adhesion molecules on cultured

human gingival fibroblasts. J Periodontol 1994; 65: 230-235.

355

Takahashi K, Poole I, Kinane DF. Detection of interleukin-1 beta mRNA-expressing

cells in human gingival crevicular fluid by in situ hybridization. Arch Oral Biol 1995;

40: 941-947.

Takahashi K, Lappin D, Kinane DF. In situ localisation of cell synthesis and

proliferation in periodontitis gingiva and tonsillar tissue. Oral Dis 1996; 2: 210-216.

Takahashi K, Mooney J, Frandsen EV, Kinane DF. IgG and IgA subclass mRNA- bearing plasma cells in periodontitis gingival tissue and immunoglobulin levels in the

gingival crevicular fluid. Clin Exp Immunol 1997; 107: 158-165.

Takashiba S, Takigawa M, Takahashi K, et al. Interleukin-8 is a major neutrophil

chemotactic factor derived from cultured human gingival fibroblasts stimulated with

interleukin-1 ß or tumour necrosis factor-a. Infect Immun 1992; 60: 5253-5258.

Tang SW, Abubakar S, Devi S, Puthucheary S, Pang T. Induction and characterisation

of heat shock proteins of Salmonella typhi and their reactivity with sera from patients

with typhoid fever. Infect Immun 1997; 65: 2983-2986.

Tanner ACR, Haffer C, Bratthall GT, Visconti RA, Socransky SS. A study of bacteria

associated with advancing periodontitis in man. J Clin Periodontol 1979; 6: 278-307.

Tanner ACR, Socransky SS, Goodson JM. Microbiota of periodontal pockets losing

crestal alveolar bone. J Periodontal Res 1984; 19: 279-291.

Tanner ACR. Is the specific plaque hypothesis still tenable? In: Guggenheim B,

editor. Periodontology Today. Basel: Karger, 1988: 123-131.

Tanner A. Microbial succession in the development of periodontal disease. In:

Hamada S, Holt SC, McGhee J, editors. Periodontal disease. Pathogens and host

immune responses. Tokyo: Quintessence publishing Co. Ltd., 1991: 13-26.

356

Tanner A, Kent R, Maiden MFJ, Taubman MA. Clinical, microbiological and

immunological profile of health, gingivitis and putative active periodontal subjects. J


Tatakis DN. Interleukin-1 and bone metabolism: a review. J Periodontol 1993;

64: 416-31.

Taubman MA, Ebersole JL, Smith DJ. Association between systemic and local

antibody and periodontal disease. In: Genco RJ, Mergenhagen SS, editors. Host-

parasite interactions in periodontal diseases. Washington D. C.: American Society for

Microbiology, 1982: 283-298.

Taubman MA, Stoufi ED, Ebersole JL, Smith DJ. Phenotypic studies of cells from

periodontal disease tissues. J Periodontal Res 1984; 19: 587-590.

Taubman MA. Eastcott JW, Schimauchi H, Takeichi 0, Smith DJ. Modulatory role of

T lymphocytes in periodontal inflammation. In: Genco R, Hamada S, Ahner T,

McGhee J, Mergenhagen S, editors. Molecular pathogenesis of periodontal disease.

Washington D. C.: ASM Press, 1994: 147-158.

Tew JG, Marshall DR, Burmeister JA, Ranney RR. Relationship between gingival

crevicular fluid and serum antibody titers in young adults with generalized and

localized periodontitis. Infect Immun 1985a; 49: 487-493.

Tew JG, Marshall DR, Moore WE, Best AM, Palcanis KG, Ranney RR. Serum

antibody reactive with preominant organisms in the subgingival flora of young adults

with generalized severe periodontitis. Infect Immun 1985b; 48: 303-311.

Tew JG, Smibert RM, Scott EA, Burmeister JA, Ranney RR. Serum antibodies in

young adult humans reactive with periodontitis associated treponemes. J Periodontal

Res 1985c; 20: 580-590.

357

Theilade E, Wright WH, Jensen SB, Löe H. Experimental gingivitis in man. II. A

longitudinal clinical and bacteriological investigation. J Periodontal Res 1966; 1: 1-

13.

Theilade E. The non-specific theory in microbial etiology in inflammatory periodontal

diseases. J Clin Periodontol 1986; 13: 905-11.

Thole ER, Hindersson P, de Bruyn J, et al. Antigenic relatedness of a strongly

immunogenic 65kDa mycobacterial antigen with similarly sized ubiquitous bacterial

common antigen. Microb Pathog 1988; 4: 71-83.

Thomas EL, Lehrer RI, Rest RF. Human neutrophil antimicrobial activity. Rev Infect

Dis 1988; 10: Supp12 S450-456.

Thompson KM, Melamed MD, Eagle K, et al. Production of human monoclonal IgG

and IgM antibodies with anti-D (Rhesus) specificity using heterohybridomas.

Immunology 1986; 58: 157-160.

Thonard JC, Crosby RG, Dalbow MH. Detection of IgM and IgA immunoglobulins in

diseased human periodontal tissue. Abstract No. 328, International Association of Dental Research: 44th general meeting, Miami, Florida 1966.

Tiebout RF, Stricker EAM, Hagenaars R, Zeijlemaker WP. Human lymphoblastoid

cell line producing protective monoclonal IgGI anti-tetanus toxin. Eur J Immunol

1984; 14: 399-404.

Tiebout RF, Stricker EAM, Oosterhof F, Van Heemstra JDM, Zeijlemaker WP.

Xenohybridisation of Epstein-Barr transformed cells for the production of human

monoclonal antibodies. Scand J Immunol 1985; 22: 691-701.

358

Tolo K, Schenk K, Brandtzaeg P. Enzyme-limked immunosorbent assay for human

IgG, IgA and IgM antibodies to antigens from anaerobic cultures of seven oral

bacteria. J Immunol Methods 1981; 45: 27-40.

Tolo K, Brandtzaeg P. Relation between periodontal disease activity and serum

antibody titers to oral bacteria. In: Genco RJ, Mergenhagen SE, editors. Host parasite

interactions in periodontal diseases. Washington: American Society of Microbiology,

1982: 270-282.

Tolo K, Schenk K, Johansen JR. Activity of human serum immunoglobulins to seven

anaerobic bacteria before and after treatment. J Periodontal Res 1982; 17: 481-483.

Toothill VJ, Van Mourik JA, Niewenhuis HK, Metzelaar MJ, Pearson JD.

Characterisation of the enhanced adhesion of neutrophil leukocytes to thrombin-

stimulated endothelial cells. J Immunol 1990; 145: 283-291.

Torabinejad M, Kettering JD, Bakland LK. Localization of IgE immunoglobulin in

human dental periapical lesions by the peroxidase-antiperoxidase method. Arch Oral

Biol 1981; 26: 677-681.

Triccas JA, Roche PW, Winter N, Feng CG, Butlin CR, Britton WJ. A 35-kilodalton

protein is a major target of the human immune response to Mycobacterium leprae.

Infect Immun 1996; 64 : 5171-5177.

Triccas JA, Winter N, Roche PW, Gilpin A, Kendrick KE, Britton WJ. Molecular and

immunological analyses of Mycobacterium aviurn homolog of the immunodominant

Mycobacterium leprae 35-Kilodalton protein. Infect Immun 1998; 66: 2684-2690.

Tsai CC, McArthur WP, Baehni PC, Evans C, Genco RJ, Taichman NS. Serum

neutralising activity against Actinobacillus actinomycetemcomitans leukotoxin in

juvenille periodontitis. J Clin Periodontol 1981; 8: 338-348.

359

Tsai CC, Shcnkcr B. 1, DiRenzo JM, Malamud D, Taichman NS. Extraction and

isolation of a lcukotoxin from Actinobacillus actinomycetemcomitans with polymixin

B. Infect Immun 1984; 43: 700-705.

Tsai CC, Ho YP, Chen CC. Levels of interleukin-1 beta and interleukin-8 in gingival

crevicular fluids in adult periodontitis. J Periodontol 1995; 66: 852-859.

Udhayakumar V, Goud SN, Subbarao B. Physiology of murine B lymphocytes I

Lifespans of anti-mu and haptenated Ficoll (thymus-independant antigen) reactive B

cells. Eur J Immunol 1988; 18: 1593-1599.

U. S. Public Health Service. National Center for Health Statistics. Periodontal disease

is adults, United States 1960-1962 PHS publication no. 1000, series 11, No. 12.

Washington DC: Government printing office, 1965.

U. S. Public Health Service NCHS. Periodontal diseases and oral hygiene among

children, United States. DHEW Publication No. (HSM) 72-1060. Vol. Series 11 No.

117. Washington DC: Government Printing Office, 1972.

U. S. Public Health Service. National Center for Health Statistics. Basic data on dental

examination findings of persons 1-75 years; United States 1971-1974. DHEW

publication no. (PHS) 79-1662, series 11, No. 214. Washington DC: Government

printing office, 1979.

U. S. Public Health Service NIDR. Oral health of United States Adults; National

findings. NIH Publ. No. 87-2868. Bethesda MD: NIDR, 1987.

Vaerman JP, Heremans JF, Laurel] CB. Distribution of achain subclasses in normal

and pathological IgA-globulins. Immunology 1968; 14: 425-

360

Van der Weijden GA, Timmerman MF, Danser MM, Nijboer A, Saxton CA, Van der

Weiden U. Fffect of pre-experimental maintenance care duration on the development

of gingivitis in a pa rtiaal mouth experimental gingivitis model. J Periodontal Res 1994;

29: 168-173.

Vandesteen GE, Williams I3L, Ebersole JL, Altman LC, Page RC. 1984.

Clinical, microbiological and immunological studies of a family with a high

prevalence of early-onset periodontitis. J Periodontol 1984; 55: 159-69.

Van Dyke TE, Schweinebraten M, Cianciola Li, Offenbacher S, Genco RJ. Neutrophil

chernotaxis in families with localized juvenile periodontitis. J Periodontal Res 1985;

20: 503-514.

Van Eden W. Flcat shock proteins in autoimmune arthritis: a critical contribution

based on the adjuvant arthritis model. APMIS 1990; 98: 383-394.

Van Mccl FCM, Stccnbakkcrs PGA, Oomen JCH. Human and chimpanzee

monoclonal antibodies. J Immunol Methods 1985; 80: 267-276.

van Winkelhoff AJ, van Steenbergen TJ, de Graaf J. The role of black-pigmented

Bacteroides in human oral infection. J Clin Periodontol 1988a; 15: 145-55.

van Winkelhoff AJ, van der Velden U, de Graaf J. Microbial succession in

recolonizing deep periodontal pockets after a single course of supra- and subgingival

debridement. J Clin Periodontol 1988b; 15: 116-122.

van Winkelhoff AJ, de Groot P, Abbas F, de Graaff. Quantitative aspects of the

subgingival distribution of Actinobacillus actinomycetemcomitans in a patient with

localized juvenile periodontitis. J Clin Periodontol 1994; 21: 199-202.

Vaquerano JE, Peng M, Chang J WC, Zhou Y-M, Leong SPL. Digital quantification of

the enzyme-linked immunospot (ELISPOT). Biotechniques 1998; 25: 830-836.

361

Vascaux R, Plot ian PA, "Lomita JK, et al. Cloning and characterisation of a new

intercellular adhesion molecule ICAM-R. Nature 1992; 360: 485-488.

Vasel D, Sims T. I, Bainbridge E3, Houston L, Darveau R, Page RC. Shared antigens of

Porphvronuonas gingivaIis and Bacteroicles forsi't/urs. Oral Microbiol Imniunol 1996;

1 1: 226-235.

Virella G. Hunioral Immune Response. Immunology Ser 1990; 50: 217-237.

Vitetta ES, Fernandcz-Botran R, Myers CD, Sanders VM. Cellular interactions in the

humoral immune response. Adv Immunol 1989; 45: 1-105.

Wakefield JD, Thorbecke GJ. Relationship of germinal centers in lymphoid tissue to

immunological memory I. Evidence for the formation of small lymphocytes upon

transfer of primed splenic white pulp to syngeneic mice. J Exp Med 1968a; 128: 153-

169.

Wakefield JD, Thorbecke GJ. Relationship of germinal centers in lymphoid tissue to

immunological memory 11. The detection of primed cells and their proliferation upon

cell transfer to lethally irradiated syngeneic mice. J Exp Med 1968b; 128: 171-187.

Waldmann TA. The regulation of the human humoral immune response and

functional assays. Birth Defects: Original Article Series 1983; 19: 31-36.

Walker C, Gordon J. The effect of clindamycin on the microbiota associated with

refractory periodontitis. J Periodontol 1990; 61: 692-698.

Wang S, Sun C, Gillanders E, et al. Genome scan for susceptibility loci to the

complex disorder early onset periodontitis. Am J Hum Genet 1996; 59: 1386 (abstr).

362

Watanabe H, Marsh PD, lvanyi L. Antigens of A. actinonrycetemcomitans identified

by imnwilohlotting with sera from patients with localised human juvenile

periodontitis and generalised severe periodontitis. Arch Oral Biol 1989; 34: 649-656.

Watanabe II, Umeda M, Seki T, Ishikawa I. Clinical and laboratory studies of severe

periodontal disease in an adolescent associated with hypophosphatasia. A case report.

J Periodontol 1993; 64: 174-180.

Weinberg A, Holt SC. Interaction of Treponema denticola TD-4, GM-1 and MS25

with human gingival fibroblasts. Infect Immun 1990; 58: 1720-1729.

Weiss J, Victor M, Stcndahl 0, Eisbach P. Killing of gram negative bacteria by

polymorphonuclcar leukocytes. Role of an 02-independent bactericidal system. J Clin

Invest 1982; 69: 959-970.

Weiss SJ. Tissue destruction by neutrophils. N Engl J Med 1989; 320: 365-376.

Weiss A, Littman DR. Signal Transduction by lymphocyte antigen receptors. Cell

1994; 76: 263-274

Wennström JL. What is a clinically healthy periodontium? Guggenheim B, editor. In:

Periodontology today. Basel: Karger, 1988: 1-5.

Wheeler TT, McArthur WP, Magnusson I, et al. Modeling the relationship between

clinical, microbiologic, and immunologic parameters and alveolar bone levels in an

elderly population. J Periodontol 1994; 65: 68-78.

Whittaker C1, Klier CM, Kolenbrander PE. Mechanisms of adhesion by oral bacteria.

Annu Rev Microbiol 1996; 50: 513-552.

363

Wicker LS, Katz M, Sercarz EE, Miller A. Immunodominant protein epitopes I.

Induction of suppression to hen egg white lysozyme is obliterated by removal of the

first three N-terminal amino acids. Eur J Immunol 1984; 14: 442-447.

Wilkinson PC. How do Ieucocytes perceive chemical gradients? FEMS Microbiol

Immunol 1990; 64: 303-312.

Williams TJ, Peck MJ. Role of prostaglandin-mediated vasodilation in inflammation.

Nature 1977; 270: 530-532.

Williams BL, Ebcrsolc . 1L, Spektor MD, Page RC. Assessment of serum antibody

patterns and analysis of subgingival microflora of members of a family with a high

prevalence of early-onset periodontitis. Infect Immun 1988; 49: 742-750.

Williams ME, Chang TL, Burke SK, Lichtman AH, Abbas AK. Activation of

functionally distinct subsets of CD4+ T lymphocytes. Res Immunol 1991; 142: 23-28.

Wilson ME. IgG antibody response of localised juvenile periodontitis patients to the

29 kilodalton outer membrane portion of A. actinomycetemcomitans. J Periodontol

1991; 62: 211-21 S.

Wilson ME, Bronson PM, Hamilton RG. Immunoglobulin G2 antibodies promote

neutrophil killing of Actinohacillus actinomvicetenicomitans. Infect Immun 1995; 63:

1070-1075.

Wilson ME, Hamilton RG. Immunoglobulin G subclass response of juvenile

periodontitis subjects to principal outer membrane proteins of Actinobacillus

actinonwcctemcomitanrs. Infect Immun 1995; 63: 1062-1069.

Wilson ME, Kalmar JR. Fc-gamma-RIIA (DC32) -a potential marker defining

susceptibility to localized juvenile periodontitis. J Periodontol 1996; 67: 323-331.

364

Wilton JMA, Bamhton JLM, Hurst TJ, Caves J, Powell JR. Interleukin-lß and IgG

subclass concentrations in gingival crevicular fluid from patients with adult

periodontitis. Arch Oral Biol 1993; 38: 55-60.

Wittwcr J W, Dickler FH, Toto PD. Comparative frequencies of plasma cells and

lymphocytes in gingivitis. J Periodontol 1969; 40: 274-275.

Wolff LF, Aeppli DM, Pihlstrom B, Anderson L, Stoltenberg J. Natural distribution of

5 bacteria associated with periodontal disease. J Clin Periodontol 1993; 20: 699-706.

Wolff LF, Koller NJ, Smith QT, Marthur A, Aeppli D. Subgingival temperature:

relation to gingival crevicular fluid enzymes, cytokines, and subgingival plaque

micro-organisms. J Clin Periodontol 1997; 24: 900-906.

Woo DS, Holt SC, Icadbcttcr ER. Ultrastructure of Bacteroides species: Bacteroides

asacchuro4vticus, Bactc'roides fi"agilis, Bacteroides melaninogenicus subspecies

melaninogc nicus and Bacte routes melaninogenicus subspecies intermedius. J Infect

Dis 1979; 139: 534-546

World Workshop Consensus Report: aggressive periodontitis. Ann Periodontol 1999;

4: 53.

Wright SD, Tobias PS, Ulevitch RJ, Ramos RJ. Lipopolysaccharide (LPS) binding

protein opsonises LPS bearing particles for the recognition of a novel receptor on

macrophages. J Exp Med 1989; 170: 1231-1241.

Wu YL, Lee LH, Rollins DM, Ching WM. Heat shock and alkaline pH-induced

proteins of ('a niplii'lohucter- jejuni: Characterisation and immunological properties.

Infect Immun 1994; 62: 4256-4260.

Wynne SE, Walsh LJ, Seymour GJ, Powell RN. In situ demonstration of natural killer

(NK) cells in human gingival tissue. J Periodontol 1986; 57: 699-702.

365

Wynne SF, Walsh LJ, Seymour GJ. Specialized postcapillary venules in human

gingival tissuc.. 1 Pcriodontol 1988; 59: 328-331.

Wyss C. Dependence of proliferation of Bacteroides forsythus on exogenous N-

acetylmuramic acid. Infect Immun 1989; 57: 1757-1759.

Wyss C, Choi BK, Schupbach P, Guggenheim B, Gobel UB. Treponema amylovorum

sp. Nov., a saccharolytic spirochete of medium size isolated from an advanced human

preiodontal lesion. ]nt J Syst Bacteriol 1997; 47: 842-845.

Yarnamura M, Uycmura K, Deans RJ et al. Defining protective responses to

pathogens: cytokine profiles in leprosy lesions. Science 1991; 254: 277-279.

Yamasaki A, Nikai H, Niitani K, Ijuhin N. Ultrastructure of the junctional epithelium

of germ free rat gingiva. J Pcriodontol 1979; 50: 641-648.

Yarnazaki K, Nakajima T, Aoyagi T, Hara K. Immunohistochemical analysis of

memory T lymphocytes and activated B lymphocytes in tissues with periodontal

disease. J Periodontal Res 1993; 28: 324-334.

Yamazaki K, Nakajima T, Hara K. Immunohistological analysis of T cell functional

subsets in chronic inflammatory periodontal disease. Clin Exp Immunol 1995; 99:

384-391.

Yarchoan R, Murphy BR, strober W, Clements ML, Nelson D. In vitro production of

anti-influenza virus antibody after intranasal inoculation with cold-adapted influenza

virus. J Immunol 1981; 127: 1958-1963.

Yavuzyilmaz E, Yamalik N, Bulut S, Ozen S, Ersoy F, Saatei U. The gingival

crevicuir fluid interleukin-1 beta and tumour necrosis factor-alpha levels in patients

with rapidly progressive periodontitis. Aust Dent Jour 1995; 40: 46-49.

366

Zachinsky LS. Range of' histologic variation in clinically normal gingiva. J Dent Res

1954; 33: 580-581).

Zachrisson BU. A histological study of experimental gingivitis in man. J Periodontal

Res 1968; 33: 580-589.

Zadeh lila, Kreutzer D. Evidence for involvement of superantigens in human

periodontal diseases: Skewed expression of T cell receptor variable regions by gingival

T cells. Oral Microbiol Immunol 1996; 11: 88-95

Zafiropoulos GG, Flores-de-Jacoby L, Schoop B, Havemann K, Heymanns J. Flow

cytometric analysis of' lymphocyte subsets in patients with advanced periodontitis. J


Zahringer U, Lindner B, Rietschel ET. Molecular structure of Lipid A, the endotoxic

centre of' bacterial lipopolysaccharides. Adv Carbohydr Chem Biochem 1994; 50:

211-273.

Zambon JJ, Slots J, Genco RJ. Serology of oral Actinobacillus

actinorni'cctcmconiituns and scrotype distribution in human periodontal disease. Infect

Immun 1983; 41: 19-27.

Zambon JJ. Actinohacillus actinonrviceteincomitans in human periodontal disease. J


Zambon J. I. Periodontal diseases: Microbial factors. Ann Periodontol 1996; 1: 879-

925.

Zamensky NH. Alveolar-pyorrhea - its pathological anatomy and its radical treatment.

Br Dent J 1902; 23: 585-604.

367

Zigmond S11. Mechanisms of sensing chemical gradients by polymorphonuclear leukocytes. Nature 1974; 249: 450-452.

Zimmerman U. Elcctric field-mediated fusion and related electrical phenomena. Biochim Biophys Acta 1982; 694: 227-277.

368

List of Publications

Abstract prescntations:

Podmore M, Lappin D, Mooney . I, Kinane DF. Microbial targets in periodontal

disease. J Dent Res 1998; 77 (Spec Iss): 1029.

Podmore M, Darby lB, Kinane DF. The effect of periodontal therapy of patients'

humoral immune response. J Dent Res 1999; 78 (Spec Iss): 137.

Review Publications:

Podmore M, Ebersole JE, Kinanc DF. Immunodominant antigens in periodontal

disease: a real or elusive concept. Crit Rev Oral Med Biol 2001; 1: (in press).

Kinanc DF, Podmorc M, Ebersole JL. Etiopathogenesis of periodontitis in children

and adolescents. Periodontology 2000 2001; 26 (in press).

369

Microbial immune targets in periodontal disease. M. PODMORE, *

D. F. LAPPIN,. l. MOONEY & D. F. KINANE. (University of Glasgow Dental

School, UK)

The immune response against periodontal pathogens is considered important in

periodontal disease. Knowledge of the microbial targets for serum and crevicular

fluid antibodies is crucial in the development of diagnostic techniques and for

immunisation strategies. The aim of the study was to examine the microbial targets of

peripheral humoral immune cells in ten patients affected by severe chronic

periodontitis (35 to 65 years of age). ELISPOT assays were used to measure the

presence of antibody producing cells in patients with periodontal disease. This assay

has provided useful preliminary data which will assist in the identification of

immunodominant antigens against which the humoral immune response in this

disease is directed. A second approach was to utilise cell culture supernatants from

Epstein-Barr virus (EBV) transformed pre-B cells from patients with periodontal

disease which were screened by ELISA for antibodies against Porphyromonas

gingivalis, Actinohacilhu actinomycetemcomitans, Prevotella intermedia and

Bacteroides forsythus. After screening 500 EBV transformed wells for each patient

we detected between 15 and 30 positive wells for each of the three putative

periodontal pathogens tested. These EBV transformed clones were combined with a

human myeloma cell line to form discrete hybridomas capable of producing

monoclonal antibodies. Thus human monoclonal antibodies can be produced from

pre-B cells of periodontitis patients and a proportion are directed against neriodontal

pathogens.

The effect of periodontal therapy on the humoral immune response in patients

with chronic periodontal disease. M. PODMORE, * I. B. DARBY & D. F.

KINANE. (University of Glasgow Dental School, UK)

Numerous bacteria and the host humoral immune response to their antigens have been

implicated in the pathogenesis of periodontal disease. The aim of this study was to

assess the effect of initial periodontal therapy on 25 untreated adult periodontitis

patients, and specifically assess the patient's serum antibody titres to a range of

bacteria and specific antigens. The antigenic targets investigated were the whole fixed

bacteria of Porphi'romonas gingivalis (P. g), Actinobacillus actinomycetemcomitans

(A. a), Prcwotc'lla intcrmedia (P. i), Treponema denticola (T. d) and Bacteroides

forsl't{rus (B. f), crude outermembrane protein preparations of P. g, A. a, P. i and B. f and

various purified antigens thought to play a role; Leukotoxin (A. a), Lipopolysaccharide

(P. g), Heat shock protein 60 (P. g) and RIA protease (P. g). Antibody titres were

assayed by ELISAs. Following the initial therapy the average reduction in pocket

depth was 2.1 mm (+ 0.9). In addition, specific antibody titres increased after

treatment for all whole bacteria and antigens tested, for example, the antibody titre to

P. g whole cells before treatment was 0.510 ELISA units (± 0.180) and after treatment

was 0.927 ELISA units (± 0.334). This increase was statistically significant (P<0.05).

Interestingly, P. g outer membrane protein preparations had a 119% increase in titre

whereas antibodies to Lipopolysaccharide from P. g increased by only 25.1%.

In conclusion, periodontal therapy influenced the magnitude of the humoral immune

response to all of the putative periodontal pathogens and specific antigens tested.

Initial periodontal treatment does not increase antibody titres to certain periodontal

pathogens and their antigens to the same extent and this may indicate the relevance of

certain candidate antigens in the disease process.

TEXT BOUND INTO

THE SPINE

IMMUNODOMINANT ANTIGENS IN PERIODONTAL DISEASE:

;A REAL OR ILLUSIVE CONCEPT?

l Podmore 1 Fbersofe ý1. Kinne*

4rersity of Glasgow Dental Hospital and School, 378 Sauchiehall Street, Glasgow, G2 311, Scotland, UK; *corresponding author, D. F. Kinane@dental. gla. ac. uk

ABSTRACT: The humoral arm of the immune system provides protection from many medically significant pathogens. The antigenic epitopes of the pathogens which induce these responses, and the subsequent characteristics of the host response, have been extensively documented in the medical literature, and in many cases have resulted in the development and implementation of effective vaccines or diagnostic tests. There is a substantial body of literature on the humoral immune response in periodontal disease, which is targeted at micro-organisms present within periodontal pockets. However, the significance and specificity of the immune response in periodontal disease have proved difficult to elucidate, due to the large number of potential pathogens in the plaque biofilm and the apparent commensal nature of many of these opportunistic pathogens. This review addresses our current knowl-

edge of the approaches and strategies which have been used to elucidate and examine the concept of immunodominant antigens in medical infections and, more recently, periodontal disease. An identification/understanding of the immunodominant antigens would be informative with respect to. (i) the relative importance of the implicated pathogens, (ii) new approaches to immunological diagnosis, (iii) specific bacterial virulence determinants, (iv) natural protective responses, and (v) the selection of potential vaccine candidate antigens. We conclude that immunodominance of antigens in periodontal disease may be relevant to our understanding of periodontal disease pathogenesis, but due to the complexity and diversity of the 'pathogenic microbial ecology', it is currently an enigmatic topic requiring a multidisciplinary approach linking clinical, microbiological, and immunological investigations. We also conclude, after assessing the literature available on the topic of immunodominance, that it is a term that, if used, must be clearly defined and understood, since it is often used loosely, leading to a general misinterpretation by readers of oral and medical literature.

Key words. Immunodorninance, periodontitis, oral pathogens, antigens, detectable antibody response.

(I) Definition of the Term "Immunodominant Antigen"

ýn dealing with this topic, we must first define the term Immunodominance'. This concept is often misused by

vestigators who go on to provide data and an interpretation

at specific antigens or epitopes of a micro-organism or ý', acromolecule are 'immunodominant', with little explana-

)n as to the basis of the finding and, importantly, its signif- I 'ance. The term 'immunodominance or immune dominance'

as initially coined by Eli Sercarz (1989) to provide a frame-

vork for describing the immune response gene control of antibody responses to hen egg lysozyme (HEL). FIEL is a , obular protein with a known three-dimensional structure, ,, hich enabled these investigators to define antigenic epi- 'ipes using peptides derived from cyanogen bromide cleav- ý5e of this protein. Specifically, immune responses of H-2h 'nice to HEL were restricted to a domain at the amino termi- 'ýus of the molecule-in fact, to 3 amino terminal residues. nicker et al. (1984) reported that if these 3 residues were '-leaved, 50% of the primary response antibody molecules ', ound poorly if at all to the HEL, and therefore, the immuno-

-,, nicity of the protein was drastically affected. The term mmunodominance' was used to describe the predominant rnmune recognition of these 3 amino terminal residues. ,, ercarz (1989) thus suggested that, for efficient and regulat-

ed activity of the immune response to a particular antigen, there should be a small number of immunodominant epitopes on an antigen.

The primary, secondary, and tertiary structures of a protein antigen present a vast array of potential epitopes, however, the number recognized by the immune system appears to be considerably smaller than the full, potentially antigenic repertoire of the immunogen (Laver et al, 1990). Some of these antigenic epitopes will dominate the resulting specificity of the immune response, which can vary from species to species and even among individuals within a species, and will not only induce a more pronounced immune response, but may also suppress the primary immune response to other antigenic determinants Recent studies with a similar experimental system have demonstrated the broad heterogeneity of proliferative responses to the immunodominant determinants within HEL. The heterogeneity of response was suggested to relate to the competitive, as well as regulatory, nature of the interaction among various factors, including different Major Histocompatibility Complex (MtiC) molecules, determinant capture by antigen-presenting mechanisms, and the available T-cell repertoire. The authors suggested that these results, while directed toward focused studies of an isolated protein, may have important implications for studies of autoimmunity, infection, and vaccine design in human populations, where heterozygosity is the characteristic of the population (Moudgil et a[_ 1998) The basis for the existence of immunodominant epitopes is less well-understood. Recent findings have

12(2) 000-000 (2001 ( Crit Rev Oral Biol Med

gun that peptide cornpelition for MI IC binding my not repre- 'the most irnpoit, int event in processes leadint; to immuno- minance (Lo-Man it of ,

I99ttf As his been noted with the ilting antibody to it particular protein antigen, T-lymphocyte ; cnses to a protein antigen are restricted to a limited number determinants and not to all peptides capable of binding to -1C class II molecules This 1orusing of the T-cell immune 'Aonse is also defined is imtnunodominance and has been : served with numerous , rntigens Recent investigations by

in et at ( 1998) hive suggested t h, it dendrit is cells play a major e in the focusing of the nnnrune response against it few anti- c determinants, while It-lymphoyte's may diversify the T-cell

'5ponse by present inl; i Hinre heterogeneous set of peptide- complexes Moreover, this "I'-cell focusing; has been extend-

to cellular immune responses, which are directed against it arrow set of irnmunoriomin, rnt peptides derived from complex 'ýtigens. The existence of epitopes that are hidden or infrequent ? gets of immune responses (cryptic epitopes) has also been 'entified. Although the identification of immunodominant epi- ? es is important for vaccine development, an understanding of ýmunologicaI reactivity to these cryptic epitopes may be impor- '? t in the development of . rutoimmunity Plum et al. ( 1997) have

Jggested that targeting exogenous antigens into specialized recessing compartments within antigen-presenting cells Spears to be important in epitope selection and immunodomi- Oce The final outcome of the immune response may be an 'tmunochemical phenomenon that is related to the protein ructure or the host's immunoregulatory mechanisms, however,

also appears to be effectively used by various pathogens as it ')st evasion tactic 'T'his occurs when the microbial immuno- 'iminant epitopes are sterically close to antigens which have the , pacity to demonstrate inl igenie diversily or are capable of anti- "nic variation Thus, they help to conceal the micro-organism

m the host-protective responses (Nara et at , I998) (AO)

In the field of periodontal research, the term 'immunodomi- unce' has been used imprecisely, and is poorly defined arid ten misrepresented-There irre multiple reports in the literature

-urportedly charrcterizintg immunodominarnt antigens of oral r'icro-organisms, yet no clear definition of what is meant by

munodornin, ince is included, arid the terra is used to describe

e detection of r high in[ibody liter in in Il, ISA, or substantial , ttibody reactivity torn anti}venue molecule that has been visu- lized on ,r Western imrnrniohlot Studies by Califano et al (1992) A l3oulsi et ir! (1'l')(ß) provide examples of investigations mis- sing the term 'Inimunodomina art antigens' in their reports of ! udies on Actinohacilluu tu ti numywti'mcumilans and I'orphyrononas

ngiva(is antibody levels in patients I)o Ihese antigens truly xpress rin c'pitopc' that is the focus of the primary immune 55spons(,, .riet which is essential for the initiation of the response ') the antigen rid micro-organism? (Wickert, ( at, 1984) The obvi- ýus answer is that currently we do not know Due to the chronic ature of periodontal infections, these objective measures may nly be defining the gravimetric quantity of a particular antigen,

li'ather than describing a function of its unique antigenicity Thus, he original definition of immunodominance has been distorted

'i describe it microbial antigen to which it predominance of anti- 't'idy is detected when compared with the myriad other antigens , resented by the inic_ro-organism and recognized by the host. For the purposes of this review, we utilize the term 'detectable anti- 'jody response' to overcome these difficulties and to avoid con- ùsion when addressin;; the periodontal humoral immune

response reports

(11) The Complexities of the Periodontitis Humoral Immune Response

Periodontal disease is a multifactorial disease with a complex polymicrobial etiology. This not only varies between various forms of clinical disease, but may also vary at the individual patient level. For example, a subject may exhibit a disease process associated with a different species or consortia of bacte-

ria at different disease sites, or recurrent disease exacerbations may occur in the context of infection and/or emergence of differ-

ent pathogenic bacterial species/consortia temporally in the site. In this regard, periodontal disease has been difficult to study, due to the large number of species comprising the microbial biofilm at sites of health and disease, proposed genetic predispositions, behavioral variables such as adoption of oral hygiene, and environmental factors (i. e., smoking, and hormonal and socio-economic factors). In addition to these factors, the nature of the humoral immune response itself is also controversial. Whether the antibody response in this disease is protective or in fact a facet of the pathology is unknown one would think that disease activity would be abrogated if the antibody response was effective at eliminating or controlling the growth of the micro-organisms present. That is of course not to say that all pathology seen in this disease is directly due to the harmful effects of micro- organisms. Production of high levels of pro-inflammatory cytokines (IL-6, IL-8, TNF) has been reported (Tsai et al., 1995; Yavuzyilmaz et al., 1995), which could result in considerable tissue destruction. In this review, however, we are dealing with the humoral antibody response in periodontal disease and hence will not discuss this issue further. A further consideration in periodontal disease is that critical clinical features of the disease are often identified a substantial period after the biologic processes have initiated the disease course.

(111) Immune Responses in Polymicrobial Infections This review addresses our current knowledge of the antigens to which a detectable immune response in periodontal disease may be directed. However, in addressing this complex issue, one must consider that within a complex microbial biofilm, such as the subgingival plaque, it appears that selected micro-organisms elicit greater antibody levels (i e., dominate the immune

response) than others (Y. ambon, 1996). This characteristic of the host response does not appear to depend entirely on the absolute numbers of an individual micro-organism, but may be

more related to their structural and antigenic components or their pathogenic significance (virulence), as well as their distribution within the biofilm and/or host tissues. An understanding of the antigens, which elicit a detectable antibody response, should contribute to a better understanding of specific bacterial

virulence determinants, natural protective responses, and even possible strategies for vaccine development, or new approaches to immunological screening and diagnosis of the disease.

(IV) Immunodominance in Microbial Diseases A survey of the scientific literature on immunodominant antigens in medical infectious diseases provides an indication of the significance of this concept for identifying potential virulence, diag-

nostic, or vaccine candidates. In the following discussion, we have reviewed carefully selected studies to provide examples of the variation in the use of the terminology, as well as the potential limitations if one equates detectable antibody response with immunodominant antigen and protective immune responses.

Cril Rev Oral ßiol Mrd 12(2) 000-000 (200'

Examples of studies identifying immunodominant and there- candidate diagnostic antigens in parasitic infections include

. lies on Neospora canium (Howe et at., 1998). Theilera parva tende et at., 1998), Trypanosoma cruzi (Pereira et at., 1998), and

1istosoma mansoni infection (Chen and Boros, 1998). It is clear '-m the literature that often an antigenic relationship to similar

gens from other parasites, and prevalence of the response t. ring infection, is used as evidence to speculate on the impor-

ce and function for potentially "important" antigens during

ection. Also, identification of important antigens from these 'es of studies can lead to a higher degree of sensitivity and ecificity in diagnostic documentation. Some studies have even , overed information on the disease pathology and progres- :n For example, Chen and Boros (1998) identified that the `rnunodominant T-cell epitope, P38, of Schistosoma mansoni elicit- the pulmonary granuloma formation, which is a disease-asso-

: ted hypersensitivity lesion of this infection. There have also been several similar studies which focused

dominance in bacterial or fungal infections. The mortality e for systemic candidosis remains high. However, the diagno- is somewhat difficult, and there has been considerable inter- in developing a reliable serodiagnostic test. Mathews et al.

88) identified the 47-kDa component of Candida albicans as an 'nmunodominant antigen' in the serology of systemic candido-

These investigators suggested that the 47-kDa antigen eems to be immunodominant, and 92% of patients with candi- mýl antibodies produced a response to an antigen of 47 kDa `4athews et al., 1984,1987). Therefore, it was proposed that this : Itigen is conserved in structure between strains, and an assay 1etecting this antigen would be the basis for a sensitive and ,, ecific diagnostic test.

Further interesting and important knowledge has been

: sned from studies involving immunodominant antigens and e quest to identify them. The Mycobacterium avium complex

1', IAC) consists predominantly of Mycobacterium avium and M. intra- %'lulare (Inderlied et al., 1993). MAC members are ubiquitous envi- `: nmental micro-organisms, and infection in healthy individuals

I: rare. However, MAC infection is a major cause of bacterial mor- "Aity and mortality in AIDS patients. A recent study by Triccas et ý' (1998) reported, after a molecular and immunological analy-

-ýs, that a 35-kDa protein of M. aviurn is a homologue of the 35-

'Da immunodominant antigen of M. leprae (Triccas et at., 1996).

ìe 35-kDa protein of MAC was confirmed to contain T-cell and B-

°II epitopes recognized by human immune responses following

., 1cobacterial infection, but it was discovered that, immunologi-

ýIly, there is an inability to distinguish between MAC and M.

+berculosis, even though the 35-kDa protein is not recognized by

Jberculosis patients. Although this is an example of a study here an antigen eliciting detectable antibody responses is iden- `led, it does not lead us to a potential vaccine candidate or any her markedly positive conclusion, For both tuberculosis and Ycobacterial infections, early diagnosis is crucial for the prog-

isis of the patients, and therefore emphasizes a situation where

-ýe use of a detectable antibody response for differential diagno-

is critical. Studies investigating dominant antigens, even if they are not

'. rimarily designed to do this, can answer crucial questions. Otitis

edia is a common problem associated with high morbidity, in

, ihich a subset of children experience repeated episodes caused

,i non-typeable Hernophilus influenzae (NTHI). Murphy and '/ungcheol (1997) have shown that the principal antibody '-, ponse in animals to an NTHI strain was directed at an

immunodominant epitope of the P2 molecule, which is the major outer membrane protein of NTHI. However, what this group hypothesized is that the otitis-prone children develop an antibody response to an immunodominant region of the P2 molecule, but that this region is strain-specific. Therefore, the child's protective response is directed exclusively at the infecting strain, and hence, he/she will experience recurrent episodes of otitis media, since he/she has no protection against a new strain. The discovery of the dominant antigen associated with this infection has further important implications, in that it contributes to our understanding of the recurrent nature of otitis media, and provides useful data on a potential vaccine design for this disease.

As mentioned in the preceding section, the importance of T- cell immunodominant epitopes is clear and must be distinguished from those eliciting a dominant B-cell response. Listeria monocytogenes secretes proteins associated with its virulence in the infected cells, which are degraded by host cell proteasomes, (AO) processed into peptides, and bound to MHC class I molecules. These antigen epitopes are effectively presented to cytolytic T- cells (CTL). Interestingly, results of Pamer et at. (1997) indicated that immunodominant T-cell responses cannot be predicted by the prevalence of antigens or epitopes. In fact, one of the least prevalent epitopes primes for the immunodominant CTL response in mice.

These are all examples of the purported significance of identification of immunodominant antigens and the potential benefits in medically important infections. Utilization of the antigen(s) to create diagnostic tools and develop vaccines attaches greater value and relevance to the search for these types of antigens.

(V) Periodontal Pathogens Concerns about identifying periodontal pathogens as true etiological agents in the disease process still exist. Many of these candidate pathogens may be merely opportunistic and depend

on the biofilm "food web" to create the conditions or niche within which they can colonize, obtain nutrients, and replicate. Thus, these micro-organisms would not classically be considered as the true pathogens to which host defense mechanisms should be directed. In spite of the potential microbial complexity of this disease, several candidate species have emerged which are thought to play a role in the pathogenesis of periodontal disease. These micro-organisms generally fulfill the modified Koch's postulates as described by Socransky and Haffajee (1992).

Examination of numerous proposed periodontal pathogens has resulted in the determination that these micro-organisms possess a variety of potential virulence factors and structures that may be crucial to their pathogenicity. In the quest for the identification of the more important bacterial antigens in periodontal disease, different research groups report, sometimes with conflicting results, different potential candidate antigens.

(A) OUTER MEMBRANE PROTEINS

Much of the periodontal literature applicable to this review topic covers studies on the Outer Membrane Proteins (OMPs) of periodontal pathogens. OMPs of Gram-negative bacteria may reside on the inner or outer layer, or can span the entire outer membrane. Gram-negative bacteria possess multiple OMPs which provide important biological functions, such as phage receptors or receptors for uptake of nutritional substrates. In this regard, various specific microbial antigens have been identified from the outer membranes of oral pathogens, and have been suggested to play a role in periodontal disease. However, many studies have

1121000-000 (2001 ý Crit Rev Oral Bio! Med

'Ified outer membrane antigens of the same micro-orga- -s, which remain uncharacterized and may be of importance ='otection afforded by the adaptive host immune response 'anabe et al., 1989; Califano et al., 1989,1992; Wilson et al., (AQ) Page et at., 1992). These studies have tended to focus : ther sonicate antigens or Outer Membrane Proteins which 3e sheared from the surfaces of the bacteria (Nakagawa et a! , Since these two preparation methods differ in the fact that produces just surface antigens and one consists of the

$ rity of the antigens of the bacteria, both internal and exter-

ýne could estimate both antibodies to intracellular antigens, , ell as to the surface antigens. Similar results, however, might expected. It seems unlikely that patients would produce an

#,; of antibodies to "inaccessible" intracellular proteins, and re likely that antigens eliciting a detectable antibody response Id be expressed on the microbial surface or as secreted pro-

's The majority of the outer membrane components of puta- periodontopathogens remain uncharacterized with respect

Heir metabolic or structural functions or their activity as viru- 'e determinants. In general, patients with strong responses to >e individual antigens often tend to be those who have a

! 'eng response to the whole bacterial cell. Nakagawa et at. (1995) reported detection of antibodies to

tarent 'immunodominant antigens' of P. gingivalis in the sera of t periodontitis patients, which were specific for antigens of 75 31 kDa molecular weight from outer membrane preparations, a 46-kDa antigen from sonicated extracts of the micro-orga-

m. Further studies identified 75-, 55-, and 43-kDa bands in '' )le-cell sonicates of P. gingivalis to be 'immunodominant anti- "'s' in rapidly progressive periodontitis patients (Chen et at., `15). A similar approach was utilized by Ebersole and Steffen

They documented 'immunodominant antigens' in outer mbrane preparations from various strains of P gingivalis. The

-, alts identified some similarities in antibody recognition of `"gen bands across strains; however, a striking variability was

ced between the strains. Additionally, variations in response 'terns were noted to be associated with high- and low-response

ý'; ents, regardless of the clinical disease classification. Ebersole et al. (1995) and others (Watanabe et at., 1989;

stad et al., 1990; Sims et al., 1991) have used Western rnunoblot approaches to detect antibody responses specific antigens in the outer membranes from A. actinornycetemcorni-

Interestingly, these studies revealed a striking variability in

ponces among individual patients, thereby posing a potential ma regarding the identification and/or importance of an anti-

that dominates throughout a patient population. 'srnatively, Flemmig et al. (1996) demonstrated IgG/IgA anti-

. diy activity in Early Onset Periodontitis (EOP) (65%/70%) and lt Periodontitis (AP) (45%/55%) patients reactive with a 110- O MP of A. actinornycetemcomitans. In contrast, control subjects no IgG, and only 5% had any IgA antibody activity. This

"proach exemplifies a strategy whereby a high frequency and/or 'I of antibody is detected to a particular antigen in patients vs.

-'troll, which is then suggested to be an immunodominant

"'igen. Each of these studies aimed to identify immunodomi-

: rit antigens as macromolecules which may play an important

1, in the pathogenesis of periodontal diseases, and to correlate

response with disease resistance. In fairness to the authors, yny of these studies probably identified the molecular weight antigens which elicited a detectable antibody response in this ease, but none of these studies obtained sufficient evidence

'lirative of the presence of an immunodominant antigen.

Importantly, the OMPs are uniformly immunogenic, and many demonstrate antigenic diversity (heterogeneity) (Holt and Bramanti, 1991, Holt and Ebersole, 1991; Mitchel, 1991). (AO) Moreover, environmental changes altering the growth of numerous micro-organisms, such as Bordetella Pertussis, have been shown to result in antigenic modulation of outer membrane proteins (Robinson et al., 1986). Genco and colleagues (1991) presented the concept of the potential importance of antigenic heterogeneity as a virulence strategy for oral micro-organisms. Recent studies by Ebersole and colleagues (1995) with A. actinomycetemcomitans, Chen

et at. (1995) with R gingivalis, and Sims et a!. (1998) with B. fürsythus have supported antigenic diversity of these pathogens within infected patients, but there remains minimal information of its

potential role in virulence within the periodontal microbiota. The previously mentioned studies are representative of

strategies which have been used and provide examples of some potentially interesting targets. However, they only "scratch the surface" in providing an understanding of the antigenic epitopes which activate the regulatory T-cells and effectively stimulate regional B-lymphocytes which could provide a functional host immunity in periodontal disease. Current genome sequencing of these pathogens will allow us to visualize the organization of these protein genes, as well as enhancing studies of genetic and antigenic heterogeneity within the population. Sequencing, purification, and functional studies of these antigens will encourage evaluation of their significance in well-characterized patient and control populations, as well as in in vitro and in vivo models of disease. More importantly, from the perspective of this review, it is not clear that the concept of a 'dominant antigen' is critical to the ultimate goals of periodontal research. As mentioned previously, high antibody levels or frequency to an antigen does not necessarily define immunodominance of the antigen, but rather delineates a dominant antibody response. If this is directed to a 'flak antigen', an antigen capable of significant diversity or variation, or having little contribution to the pathogenicity of the micro-organism, the detection of a dominant antibody response does not inherently describe an immunodominant antigen or provide insight into immune-protective responses

(B) FIMBRIAE

The available literature supports that fimbriae of P. gingivalis are a target of the humoral immune response in periodontal disease patients, although a lack of antibodies reactive with fimbriae in patients has not been defined as increasing disease risk. This has been delineated in various disease populations, with a general finding of increased levels and frequency of serum antibody to the fimhriae or fimbrillin subunits (Okuda et al., 1988, (AO) Ogawa et al., 1989, Genco e( al., 1991 ). Recently, Condorelli et al. (1998) reported the detection of IgA antibodies in the gingival crevicular fluid (GCF) of patients with acute recurrent periodontitis. The antibodies were specific for the purified 43-kDa fimbrial antigen. Thus, it appears clear that both a local and a systemic humoral immune response are generated to this antigen in many periodontitis patients. Evidence that firnbriae are immunodominant

antigens in periodontal disease, or in fact in other infectious dis-

eases, is generally lacking.

(C) LIPOPOLYSACCHARIDE

Numerous studies have demonstrated that the serotype deter-

minants of Aa ctinomycetemcomitans are associated with a high-

molecular-weight LPS-associated antigen (Califano et al , 198 9)

or carbohydrate smear antigen (Califano et al., 1989, Sims et al,

Cril Rev Oral E3ioI Med 12(2)-000-000 (2001)

'; i. Interestingly, in Early Onset Periodontitis (EOP) patients the highest serum antibody levels to A. actinornycetemcomi- the predominant antibody specificity is to the LPS (Okuda

al, 1986; Wilson, 1991; Califano et al., 1992, Wilson and rnilton, 1995). Moreover, while LPS has generally been aught to exhibit immune-stimulating characteristics which

T-cell-independent, and thus induce principally an IgM

aonse, it is clear that EOP patients produce substantial lev-

of IgG antibody to this antigen from A. actinomycetemco nitans et al., 1993). This antibody also demonstrates apparent sub-

: )s specificity, in that it is primarily/exclusively of the IgG2

. class (Lu et al., 1993). The level and characteristics of this iinant antibody response suggest that an anti-LPS response significant marker of A. actinomycetemcomitans infection and

sting periodontitis. However, whether the LPS represents a 'Jlence determinant, a true immunodominant antigen, or an 'igen to which the adaptive immune response can provide . tective immunity remains to be delineated.

The literature is replete with studies reporting a significant -valence of periodontal disease patients with high titers of '+um IgG antibodies to P gingivalis, although the temporal acqui- on of the antibody response relative to colonization and/or the

èctious processes associated with this micro-organism have ý. t been clearly delineated. Thus, strategies involving animal Aels have been developed to examine the acquisition and Qcificity of potential protective immunity. A recent study by

Esel et al. (1996) reported that immunization of monkeys with a iccine containing a monkey isolate of P. gingivalis induced pro- 41tion against alveolar bone destruction.

Consequently, the validity of targeting LPS from oral

'dhogens as an immunodominant antigen from the perspective identifying naturally protective responses for the purpose of

evention or intervention through vaccine implementation is il fraught with critical unknowns. Nevertheless, it appears that, least with certain periodontopathogens, the detectable anti-

'ody response to these types of antigens may provide some use-

.I diagnostic information.

(D) HEAT-SHOCK PROTEINS

. '! tidies supporting the importance of Heat-shock Proteins have

èn reported in the periodontal literature for some time.

munoblot analyses with purified P gingivalis GroEL revealed at it was recognized by antibody in 8/10 sera from patients test-

and in only 3/9 healthy controls (Maeda et al., 1994). Evidence

this nature, however, does not warrant a further discussion of 'is antigen in a review looking at immunodominance.

(E) LEUKOTOXIN

'=riodontitis patients with A. actinomycetemcomitans infections

'nerally demonstrated extreme elevations in antibody to

"ikotoxin (Califano el al., 1989; Ebersole et al., 1991; Gunsolley

al_ 1991), suggesting that it is a strong immunogen. This has

confirmed in both non-human primate (unpublished

')servations) and rodent models (Shenker et al., 1993). Details 4en

the human response consistently showed elevated leukotox-

antibody in localized early-onset periodontitis (LEOP)

-7tients (McArthur et al, 1981; Tsai et al., 1981; Ebersole of a!., ')831. Relative to the focus of this review, a crucial point is that

antibody to the leukotoxin was not a dominant antibody `sponse in these patients. As mentioned previously, the mag- tude of response was clearly to the LPS and serotype determi-

nants from A. actinomycetemcomitans. The resulting interpretations suggested that LPS was an immunodominant antigen, and thus the response to this molecule should be the focus of immunological efforts to interfere with its pathogenicity. Although the biologic relevance of the leukotoxin as a virulence factor has not been clearly demonstrated in vivo, recent genotypic studies support the concept that A. actinomycetemcornitans isolates which have an altered promoter region, resulting in enhanced leukotoxin production, are most frequently associated with disease

sites in selected populations (DiRenzo et al., 1994; (AO) Haubek et at., 1995). Additionally, analysis of recent data suggests that leukotoxin antibody levels may be more discriminatory than antibody to the whole bacteria (Califano et at., 1997). Furthermore, limited evidence from animal models has suggested that challenge by leukotoxin, of a human immune system reconstituted into mice, led to the induction of antibodies which correlated with some protection from lesion formation (Shenker et al, 1993). Therefore, while the reactivity of humans to the leukotoxin cannot be considered a 'dominant' response, there appear to be 'dominant epitopes' on this antigen which can consistently elicit antibody in infected subjects that appears(AO) to ameliorate the effects of this toxin. Consequently, the use of antibody to the leukotoxin as a bio-

marker for this disease, and as an immunological target, seems worthy of continued investigation.

(VI) Concluding Remarks As mentioned previously, the examples in this review are to provide an overview of the approaches and strategies which have been used in medical infections and are currently being used in

periodontal infections to examine the concept of immunodomi-

nant antigens. (AO) Where does the future lie for the immu-

nodominance concept in periodontal disease research? 'Immunodominance' in periodontal disease may be a relevant entity. However, with the complexity of this polymicrobial infection and the intertwining of the relationships within the microbial ecology, the plethora of host and microbial variables may make any successful exploitation of any intervention strategy based on the immunodominance concept difficult to implement in this dis-

ease. This concern is exemplified by the breadth of studies supplying often-disparate findings regarding the antigenic specificity of dominant host responses to the group of putative pathogens

One important consideration emphasized in this review is that future studies and published reports must be more careful in their use of the term 'immunodominance'. Authors should be clear what they mean by this term and must define it for their readers. 'Immunodominance' is a term too readily used in an unquali- fied manner, leading to much confusion and misinterpretation.

If the identification of immunodominant antigens is relevant and additional supportive knowledge is gained, we would suggest that utilization of these critical antigens for screening and diagnosis of the disease, as well as using them as targets for

vaccine development, will be not only feasible but also desir-

able. This is a rich area for future periodontal disease research, and we anticipate that additional data will more accurately identify whether critical antigens are immunodominant or are just the target of a detectable antibody response. Importantly, this area of research will require multidisciplinary expertise from "chairside to lab bench" in linking clinical, microbial, and immunological knowledge and techniques to elucidate this as- yet-elusive target.

tl2) 000-00)) (2001) Crit Rev Oral Biol Med

QFERENCES(AQ)

. ý"n IS, Ma C, Kovats S( 1997). Antigen-presenting cells and the selection of immunodominant epitopes. Crit Rev Immunol 17411-417. stad Al, Kristoffersen T, Olsen I, Preus HR, Jensen HB, Aasstrand EN, et at. (1990). Outer membrane proteins of ýctinobacillus actinomycetemcomitans and Haernophilus aphrophilus studied by SDS-PAGE and immunoblotting. Oral Microbial Immunof 5: 155-161. Jtsi EA, Koseki T, Nishihara T, Ishikawa I (1996). Characterisation of the immunodominant antigens of Porphyromonas gingivalis 381 in high-responder patients. Oral tvticrobiol Immunol 11: 236-241.

. fano IV, Schenkein HA, Tew JG (1989). Immunodominant antigen of Actinobacillus actinomycetemcornitans Y4 in high-responder

patients. Infect Immun 57: 1582-1589. ! fano IV, Schenkein HA, Tew IG (1992). Immunodominant antigens of Actinobacillus actinomycetemcornitans serotype b in early onset periodontitis patients. Oral Microbial Immunol 7: 65-70.

: jifano IV, Pace BE, Gunsolley IC, Schenkein HA. Lally ET, Tew IG 11997). Antibody reactive with Actinobacillus actinomycetemcornitans leukotoxin in early-onset periodontitis patients. Oral Microbiol Immunol 12: 20-26.

en HA, Weinberg A, Darveau RP, Engel D, Page RC (1995). Immunodominant antigens of Porphyromonas gingivalis in

j patients with rapidly progressive periodontitis. Oral Microbial Immunol 10: 193-201.

en Y, Boros DL (1998). Identification of the immunodominant T cell epitope of p38, a major egg antigen, and characterization of the epitope-specific Th responsiveness during murine Shisotsomiasis(AO) mansoni. I Immunol 160: 5420-5427.

': ndorelli F, Scalia G, Cali G, Rossetti B, Nicoletti G, Lo Bue AM

(1998). Isolation of Porphyromonas gingivalis and detection of immunoglobulin A specific to fimbrial antigen gingival crevicular fluid. I Clin Microbial 36: 2322-2325.

Renzo(AO) IM, McKay TL (1994). identification and characterisation of genetic cluster groups of Actinobacillus actinomycetemcornitans isolated from the human oral cavitiy. I Clin Microbial 3275-81.

'ýersole IL, Steffen MJ (1994). Human antibody responses to

outer envelope antigens of Porphyromonas gingivalis serotypes. Periodont Res 30: 1-14.

ersole IL, Taubman MA, Smith DI, Hammond BE Frey DE

(1983). Human immune responses to oral micro-organisms. U. Serum antibody responses to antigens from Actinobacillus actinomycetemcomitans and the correlation with localised juvenille

periodontitis. I Clin Immunol 3: 321-331.

ersole JL, Sandoval M-N, Steffen MI, Cappelli D (1991). Serum

antibody in Actinobacillus actinomycetemcomitans-infected patients

with periodontal disease. Infect Immun 59: 1795-1802.

', ersole JL, Cappelli D, Sandoval M-N, Steffen MI (1995). Antigen

specificity of serum antibody in Actinobacillus actinomyceterncomitans infected periodontitis patients. I Dent Res 74,658-666.

-mmig TF, Selmair I. Schmidt H, Karch H (1996). Specific antibody reactivity against aI l0-Kilodalton Actinobacillus actino-

mycetemcomitans protein in subjects with periodontitis. Clin

Diagn Lab Immunol 3678-681 .

-, pin L, Bravo de Alba Y, Casrouge A, Cabaniols )P, Kourilsky P, Kanellopoulos I (1998). Antigen presentation by dendritic

cells focuses T cell responses against immunodominant peptides, studies in the hen egg-white lysozyme (HEL) model. I Immunol 160: 1555-1564.

ý, nco RJ, Lee I-Y, Sojar HT, Bedi GS, Loos BC, Dyer DW (1991). Antigenic heterogeneity of periodontal pathogens. In

Periodontal disease: pathogens and host immune responses. Tokyo: Quintessence Publishing co., Ltd., pp. 167-183.

Gunsolley JC, Tew JG, Connor T, Burmeister JA, Schenkein HA (1991). Relationship between race and antibody reactive with periodontitis-associated bacteria. I Periodont Res 26: 59-63.

Haubek D, Poulsen K, Asikainen S, Kilians M(AO) (1995). Evidence for absence in Northern Europeans of especially virulent clonal types of Actinobacillus actinomycetemcomitans. I Clin Microbial 33395-401.

Holt SC, Bramanti TE (19911. Factors in virulence expression and their role in periodontal disease pathogenesis. Crit Rev Oral Biol Med 2: 177-281.

Holt SC, Ebersole IL (1991). The surface of selected periodontopathic bacteria: possible role in virulence. In: Periodontal dis-

ease pathogens and host immune responses. Hamada S, Holt SC, McGhee JR, editors. Tokyo: Quintessence Pub. Co., Ltd.,

pp. 79-98. Howe DK, Crawford AC, Lindsay D, Sibley LD (1998). The p29 and

p35 immunodominant antigens of Neospora caninum tachyzoites are homologous to the family of surface antigens of Toxoplasma gondii. Infect Immun 66: 5322-5328.

Inderlied CB, Kemper CA, Bermudez LEM (1993). The Mycobacterium avium complex. Cfin Microbiol Rev 6266-310.

Katende J, Morzaria S, Toye P, Kilton R, Nene V, Nkonge C, et al (1998). An enzyme-linked immunosorbent assay for detection

of Theileria parva antibodies in cattle using a recombinant polymorphic immunodominant molecule. Parasitol Res 84: 408-416.

Laver WG, Air MN, Webster RG, Smith-Gill SJ (1990). Epitopes on protein antigens: misconceptions and realities. Cell 61: 553- 556.

Lo-Man R, Langeveld JR Martineau P. Hofnung M, Meloen RH. Leclerc C (1998). Immunodominance does not result from

peptide competition for MHC class II presentation. I Immunol 160: 1759-1766.

Lu H, Califano IV, Schenkein HA, Tew JG (1993). Immunoglobulin

class and subclass distribution of antibodies reactive with the immunodominant antigen of Actinobacillus actinomycetemcomifans serotype b. Infect Immun 61 2400-2407.

Maeda H, Mijamoto M, Hongyo H, Nagai A, Kurihara H, Muraijama(AQ) Y (1994). Heat shock protein 60 (GroEL) from Porphyromonas gingivalis: molecular cloning and sequence analysis of its gene and purification of the recombinant protein. FEMS Microbiol Lefts 119129-136.

Mathews RC, Burnie IP, Tabaqchali S (1984). Immunoblot analysis of the serological responses in invasive aspergillosis. I Clin Pathol 38: 1300-1303.

Mathews RC, Burnie P. Tabagchali S (1987). Isolation of immunodominant antigens from sera of patients with systemic candidosis and characterisation of serological response to Candida

albicans. I Clin Microbiol 25 230-237. Mathews R, Wells C, Burnie )P (1988). Characterisation and cellu-

lar localisation of the immunodominant 47-kDa antigen of Candida albi(ans. I Med Microbiol 27: 227-232.

McArthur WP, Tsai CC, Baehni PC, Genco R), Taichman NS (1981). Leukotoxic effects of Actinobacilfus actinomycetemcomitans modulation by serum conponents. I Periodont Res 16: 159-170_

Mitchel GF(AO) (1991). Co-evolution of parasites and adaptive immune responses. Immunol Today 22. A2-A5.

Moudgil KD, Wang ), Yeung VP, Sercarz EE (1998). Heterogeneity

of the T cell response to imrnunodominant determinants

within hen eggwhite lysozyrne of individual syngeneic hybrid

FI mice implications for autoimmunity and infection. I Immunol 16116046-6053.

Murphy TF, Kyungcheol YI (1997). Mechanisms of recurrent otitis

media: importance of the immune response to bacterial sur-

Crit Rev Oral ßiol Med 12(2) 000-000 ( 2(l) I)

ace antigens Ann NY Acad Sci 810: 35'3-'360_ "igawa T, Nakagawa S, Ishihara K, Yarnada S, Machida Y, Okuda K 1995). Reactive antibodies in sera from pubertal and adult ingivitis patients against various Porphyromonas gingivalis anti-

; ens. I Periodont Res 30: 396-403. a PL, Garrity R (1998). Deceptive imprinting: a cosmopolitan strategy for complicating vaccination Vaccine I' I780-1787.

, va T, McGhee ML, Moldoveanu M (1989). Bacteroides fimbri-

se-specific IgG and IgA subclass antibody secreting cells iso- ated from chronically inflamed gingival tissues. Clin Exp

mmunol 76: 103-110. )'. da K, Takazoe I (1988) The role of Bacteroides gingivalis in perio-

Jontal disease. Adv Dent Res 2: 260-268

, da K, Kato T, Naito Y, Takazoe I( 1986) Precipitating antibody against lipopolysaccharide of I laemophilus actinomycetemcomitans n human serum. I Clin Microbiol 24846-848_

'; e RC, Sims TI, Moncla DI, Bainbridge RP, Engel LD (1992). Reactive antigens of periodontopathic bacterium A actinornycetemcomitans. Adv Exp Med Biol 327: 243-253.

ler EB, Sijts Al, Villaneuva MS, Busch DH, Vijh S (1997). MHC

class I antigen processing of Listeria monocytoqenes proteins: mplications for dominant and subdominant CTL, responses. Immunol Rev 158129-136.

"eira CM, Yamaauchi(AQ) LM, Levin MI, da Silveira IF, Castilho BA (1998). Mapping of B cell epitopes in an immunodominant antigen of Trypanosorna cruzi using fusions to the Escherichia coli LamB protein. FEMS Microbial Lett 164: 125-131.

f binson A, Duggleby Cl, Gorringe AR, Livey 1 (1986). Antigenic variation in Bordetella pertussis. In: Antigenic variation in infectious disease. Birbeck TH, Penn CW, editors. Washington, DC: IRC Press, pp. 147-161.

ýrcarz EE (1989). The architectonics of immune dominance: the aleatory effects of molecular position on the choice of antigenic determinants. Chem Immunol 46169-185.

enker BI, Wannberg S, Vitale L, Kieba I, Lally ET (1993). Reconstruction of severe combined immunodeficient mice with lymphocytes from patients with localised juvenile periodontitis. I Periodont Res 28 487-490.

'ns T), Moncla BL Darveau RP, Page RC (1991). Antigens of Actinobacillus actinomycetemcomilans recognised by patients with uvenille periodontitis and periodontally normal subjects. Infect Immun 59913-924.

rns TI, Mancl LA, Braham PH, Page RC (1998) Antigenic variation in Bacteroides forsythus detected by a checkerboard enzyme-

linked immunosorbent assay. Clin Diagn Lab Immunol 5.725-731 Socransky SS, Haffajee AD (1992) The bacterial etiology of

destructive periodontal disease: current concepts. I Periodontol 63 322-331.

Trittas IA, Roche PW, Winter N, Feng CG, Butlin CR, Britton W) (1996). A 35-kilodalton protein is a major target of the human immune response to Mycobacterium leprae. Infect immun 64: 517I- 5177.

Trittas )A, Winter N, Roche PW, Gilpin A, Kendrick KE, Britton W) (1998). Molecular and immunological analyses of Mycobacterium avium homolog of the immunodominant Mycobacterium leprae 35-Kilodalton protein. Infect Immun 66: 2684- 2690.

Tsai CC, McArthur WP, Baehni PC, Evans C, Genco R), Taichman NS (1981). Serum neutralising activity against Actinobacillus actinomycetemcomitans leukotoxin in juvenile periodontitis. I Clin Periodont 8: 338-348.

Tsai CC, Ho YP, Chen CC (1995). Levels of interleukin-I beta and interleukin-8 in gingival crevicular fluids in adult periodontitis. I Periodontol 661852-859.

Vasel D, Sims TI, Bainbridge B, Houston L, Darveau R, Page RC ( 1996). Shared antigens of Porphyromonas gingivalis and Bacteroides forsythus. Oral Microbiol Immunol 1 1226-235.

Watanabe H, Marsh PD, Ivanyi L (1989). Antigens of A. actinomycetemcomitans identified by immunoblotting with sera from patients with localised human juvenile periodontitis and generalised severe periodontitis. Arch Oral Biol 34: 649-656.

Wicker LS, Katz M, Sercarz EE, Miller A (1984). Immunodominant protein epitopes I. Induction of suppression to hen egg white lysozyme is obliterated by removal of the first three N-terminal amino acids. Eur I Immunol 14: 442-447.

Wilson ME (1991). IgG antibody response of localised juvenile periodontitis patients to the 29 kilodalton outer membrane portion of A. actinomycetemcomitans. I Periodontol 62: 211-2 18.

Wilson ME, Hamilton RG (1995). Immunoglobulin G subclass response of juvenile periodontitis subjects to principle outer membrane proteins of Actinobacillus actinomycetemcomitans. Infect Immun 63: 1062-1069.

Yavuzyilmaz E, Yamalik N, Bulut S, Ozen S, Ersoy F, Saatchi U (1995). The gingival crevicular fluid interleukin-I beta and tumour necrosis factor-alpha levels in patients with rapidly progressive periodontitis. Aust Dent 140: 46-49.

Zambon 11 (1996). Periodontal diseases: microbial factors. Ann Periodontol 1: 879-925.

2Q) 000-000 (2001) Crit Rev Oral Biol Med 7

PERD017881 P788 26-10-00 12.38 35

Periodonrofogy 2000, Vol. 26,2000.06-O() Printed in Denmark All rights reserved

20- disk/mona / a64788/pe2: 26: 04

Copyright © Munksgaard 2000

PERIODONTOLOGY 2000 ISSN 0906-6713

Etiopathogenesis of periodontitis in children and adolescents DENIS F. KINANE, MICHELLE PODMORE & JEFFREY EBERSOLE

This chapter provides an account of knowledge on the etiopathogenesis of the various periodontal dis-

eases that affect children and adolescents. This is

currently an area of active research, and many challenging questions remain. In many cases we have in-

sufficient knowledge to be definitive about the processes involved. This chapter describes the processes pertinent to the periodontal diseases, with particular emphasis on the inflammatory and immune responses occurring in the periodontal region during

gingivitis and periodontitis. This chapter will be structured in the following

way: firstly the basic elements of the host response processes relevant to periodontal disease will be related; then follows a description of how the processes interact in the pathogenesis of the diseases;

and lastly the specifics of the host response which may be relevant to the causation and diagnosis of the various forms of periodontal disease will be considered.

As described in the first chapter in this volume, the periodontal diseases affecting children and adolescents are numerous. These diseases can be con- veniently grouped into the following entities: gingivitis; early-onset forms of periodontitis; necrotizing gingivitis and periodontitis; incipient adult periodontitis; and periodontitis associated with systemic diseases. Further categories exist within these broad definitions, in particular, gingivitis has many forms, but in children the puberty-related form is common and most significant. Similarly, early-onset periodontitis has several categories namely: prepubertal, which can be localised or generalized; early-onset periodontitis, which again can be localised or generalized; and incidental early-onset periodontitis (not

extensive enough to fit any other category). All of the early-onset periodontal diseases are initiated by

dental plaque and result in destructive disease pro-

gression in susceptible individuals. The susceptibility criteria for these diseases remain elusive; the im-

mune causation could be due to variations or defects

in the host response to the plaque, however, the immune and inflammatory responses during the development and progression of periodontitis are highly complex, making inferences more difficult to draw. For example, reduced neutrophil chemotaxis has been reported in localized early-onset periodontitis and generalized early-onset periodontitis (93,94, 319,320,323-325) but is not a consistent finding

across all populations (152,153). If the abnormalities actually exist, they do not appear to predispose the patient to other diseases, suggesting that these are either mild abnormalities or only relevant in the periodontal environment (152,153).

In the many cases where periodontal disease is

predisposed to by systemic disease, identifiable

components of the immune or inflammatory process may be impaired or absent. In leukocyte adhesion deficiency or neutropenia, the systemic defect is obvious. The effect of the absence of this component in the host response can be deduced from

the description of the normal immune and inflam-

matory responses in periodontal disease. The pathogenesis of the periodontal destruction in the forms

related to systemic diseases is covered in the chapter by Gonzalez & Meyle (203) in this volume.

A pertinent aspect when considering periodontal disease in children is the hormonal influences on the tissues and the host response, which are particularly relevant in pubertal gingivitis. In addition, the known variations in the host response in early-onset periodontitis are described, but the genetics of this

and other periodontal diseases are dealt with fully in

a further chapter in this volume. Briefly, early-onset periodontitis is clearly a familial disease with an autosomal dominant mode of inheritance. Genes

coding for several aspects of the host immune response have been suggested as candidate markers to be investigated in early-onset periodontitis. These have included allelic variations in the Fc receptor for immunoglobulin G-, (264,343). The interleukin-1 polymorphisms associated with adult periodontitis

1

Kinane et al.

have not, however, been found in the early-onset periodontitis patients studied and further may suggest intrinsic differences between these disease entities.

The early-onset forms of periodontal disease most probably share a common pathogenic pathway, which may, in fact, be similar to that of adult periodontitis, although the causal pathways may differ

considerably. The predisposing aspects responsible for the much earlier onset and more rapid destruc-

tion in early-onset forms of periodontitis are probably genetically related but are still not fully understood. Thus, the pathogenesis of adult periodontal disease is covered in some detail and points relevant to children and adolescents are highlighted. The etiopathogenesis of necrotizing forms of periodontitis are quite unique and are also discussed.

The host defenses in periodontal diseases

The host defense system comprises a collection of tissues, cells and molecules whose function is to

protect the host against infectious agents (275). The immune response may be subdivided into two broad divisions, the innate (nonspecific) and the

adaptive (specific) responses. Innate reactions in-

clude the inflammatory response and do not in-

volve immune mechanisms. Adaptive or immune

responses tend to be more effective, as the host response is a specific immune response to the offending pathogens.

Innate immunity represents an important first line

of defence against infectious agents. This type of im-

munity is present from birth, is not enhanced by

prior exposure and lacks memory. The innate response has the advantage of speed, but lacks specificity and may cause host tissue damage. Innate im-

munity entails a number of elements, both cellular and noncellular. Physical barriers such as the skin and mucous membranes represent a component that infectious agents must breach to gain access to the host. The washing action of fluids such as tears,

saliva, urine and gingival crevicular fluid keeps mucosal surfaces clear of invading organisms and also contain bactericidal agents.

The intact epithelial barrier of the gingiva, sulcular and junctional epithelium normally prevents bac-

terial invasion of the periodontal tissues. It is normally an effective physical barrier against bacterial

products and components. The epithelial cell wall,

secreted proteins and fatty acids are toxic to many microbes. Salivary secretions provide a continuous flushing of the oral cavity as well as providing a continuing supply of agglutinins and specific antibodies. Furthermore, the gingival crevicular fluid flushes the gingival sulcus and delivers all the components of serum, including complement and specific antibodies.

The microbial biofilm that colonizes the tooth surface releases large quantities of metabolites that may diffuse through the junctional epithelium. These metabolites include fatty acids, peptides and lipopolysaccharides of gram-negative bacteria. By releasing proteolytic and noxious waste products, plaque microorganisms may damage cellular and structural components of the periodontium. As well as releasing these waste products, microorganisms could also invade the soft tissues.

Microorganisms produce a large variety of soluble enzymes in order to digest extracellularly host proteins and other molecules, thereby producing nutrients for growth. They also release numerous metabolic products, such as ammonia, indole, hydrogen

sulfide and butyric acid. Among the enzymes released by bacteria are proteases capable of digesting

collagen, elastin, fibronectin, fibrin, and other components of the intercellular matrix of epithelial and connective tissues. The released waste products stimulate junctional epithelial cells to release various inflammatory mediators including interleukin-1,

prostaglandin EZ and matrix metalloproteinases, all of which can transverse the junctional epithelium and enter the crevicular fluid (1).

The normal flora of the body can also act as an effective buffer against infection, by inhibiting the growth of pathogenic organisms by competition for

nutrients or production of inhibitors. Phagocytic

cells in the blood stream and tissues can destroy in-

vading agents. These include polymorphonuclear leukocytes (or neutrophils), monocyte/macrophages, and natural killer cells. Finally, there are the soluble components. These consist of a wide range of molecules that are normally protective, but, how-

ever, can also damage microbial cell walls, aid phagocytosis and cell recruitment or prevent cellular infection. These soluble components include lyso-

zymes, antimicrobial peptides, cytokines, acute- phase proteins, complement components and Inter- ferons.

The persistence of an infection in spite of the actions of the innate immune response leads to the induction of an adaptive immune response. Adaptive immune responses are characterized by 1) specificity

2

Etiopathogenesis of periodontitis in children and adolescents

for the offending antigen(s), 2) memory, which allows a more rapid and heightened response upon reinfection by the same or closely related antigen, 3) diversity, the ability to respond to a wide range of different antigens, and 4) self versus non-self recognition. The adaptive immune response can be subdivided into humoral and cell-mediated immunity. Humoral immunity is mediated by antibodies, whereas cell-mediated immunity involves the direct

action of immune cells. The lipopolysaccharides of gram-negative micro-

organisms are capable of invoking both the inflammatory and immune responses as it interacts

with host cells. The immune response will result in further release of cytokines and proinflammatory mediators, which in turn will increase the inflam-

mation and thus be more harmful to the host. The

quality of the host inflammatory response is critical to the disease process. Although its purpose is protection and prevention of bacterial invasion into the tissues, it is also detrimental, as, it can be an ineffec-

tive, chronic and frustrated response that actually causes much of the tissue damage that occurs in

periodontal disease. Inflammation, acute and chronic, nonspecific and immune-mediated, all play a role in periodontal disease. The components of these responses are now considered.

Inflammatory mediators Inflammatory mediators play a major role in acute

and chronic inflammation, and there is a strong evidence for participation of these mediators in periodontitis. These comprise a range of interacting

molecules that include the cytokine system, proteinases, proteinase activators, thrombin, histamine,

prostaglandins, leukotrienes, tissue and blood fac-

tors such as Hageman's factor, complement and clotting factors. The purpose of these and many other mediators is to initiate and perpetuate and eventually terminate an inflammatory response to insult.

In the periodontium they are produced by activated

resident gingival cells and infiltrating leukocytes. In

the blood plasma they are produced by the comple-

ment cascade and kinin system. Monocytes from in-

dividuals susceptible to or suffering from severe

periodontitis produce elevated amounts of mediators (87). Mediators are present in inflamed gin-

giva and gingival crevicular fluid from diseased sites in high concentrations (221). Concentrations of in-

flammatory mediators typically decrease following

successful periodontal therapy. There now follows a

discussion of the various components capable of promoting, sustaining and terminating inflammatory and immune reactions.

Cytokines

The term "cytokine" is derived from the Greek words kytos meaning cell, and kinesis, meaning movement. Cytokines are low-molecular-weight polypeptides of (5-70 kDa) (17). They function as soluble mediators produced by cells, to regulate or modify the activity of other cells. The cytokines can be broadly split into two groups, those involved in inflammatory and immune reactions and those involved in tissue growth and repair (growth- and colony-stimulating factors).

The first group is further subdivided into:

" the interleukins transmit information between leucocytes;

" the chemokines (chemotactic cytokines) involved in cell recruitment; and

" the interferons that influence lymphocyte activity.

Various cells secrete cytokines in response to injury

or stimulation. These proteins are biologically active in even femtomolar concentrations. Most cytokines are multifunctional molecules that act locally and have a variety of target or effector cells (341). Cyto- kines act by binding to specific receptors on the surface of effector cells, which then stimulate intracellular activation of the cell. In certain situations receptors become saturated and cytokines such as tumor

necrosis factor, interleukin-1 and interleukin-6 may spill over into the systemic circulation and stimulate distant tissues and organs (216).

Collectively, the cytokines and other related molecules form a complex network that controls the inflammatory and immune responses (356) and subsequent tissue healing. Many cytokines have overlapping functions, and some may inhibit the action of others (156). A cytokine may have different forms

of receptors that may produce opposing actions, and paradoxically, elevated levels of proinflammatory cytokines may be evident during periods of healing

as well as during destructive phases. These are im-

portant considerations when interpreting periodontal studies of single or even combinations of cytokines. These studies may not reflect the true situation in the living organism (356).

Cytokine regulation is achieved through several mechanisms. Control of cytokines can take place at

3

Kinane et al.

the gene activation level, during secretion and circulation and at the cell receptor level (17). Cytokine

production is usually short-lived and self-limiting (137). In addition, many cytokine receptors exist in

soluble forms, which may be cleaved from the target

cells, allowing them to bind and neutralize cytokines extracellularly (17). Thus, the local tissue environment influences cytokine and cytokine receptor activity. Proinflammatory cytokines are also regulated by high-affinity autoantibodies and anti-inflammatory cytokines such as interleukin-1 receptor antagonist, interleukin-4 and interleukin-10 (17).

Interleukin-1 In humans, two distinct interleukin-1 gene products have been cloned interleukin-la and interleukin-1ß. Sources of interleukin-1 production includes mononuclear phagocytes, keratinocytes (186), fibroblasts (126), endothelial cells (204) and osteoblasts (120). Interleukin-1 is a major mediator in periodontitis (235,289). Production is induced by lipopolysac-

charide and other bacterial components and by in-

terleukin-1 which is autostimulatory (216). Interleu- kin-1 is mainly secreted by monocytes or macrophages but can also be secreted by most nucleated cells (144,197).

All three members of the interleukin-1 group (in-

terleukin-la, interleukin-lß and interleukitt-1 receptor antagonist) bind to interleukin-1 receptors I and interleukin-1 receptors II. Corticosteroids induce in-

terleukin-1 receptors II messenger RNA synthesis and the release of secreted interleukin- 1 receptors II,

mostly from neutrophils, which may partly explain their anti-inflammatory and immunosuppressive actions (17).

Tumor necrosis factor

Tumor necrosis factor-a is a multipotential cytokine, produced mainly by macrophages, with a wide variety of biological effects similar to interleukin-1 (43, 170). Tumor necrosis factor-a and interleukin-1 both

act on endothelial cells to increase recruitment of polymorphonuclear leukocytes and monocytes to

sites of inflammation (20). Tumor necrosis factor-a

and interleukin-I are key mediators of chronic inflammatory diseases and have the potential to initiate tissue destruction and bone loss in periodontal disease (22). Tumor necrosis factor also mediates connective tissue destruction through its action on the matrix metalloproteinase system (51,200). Tu-

mor necrosis factor receptors exist in membrane-

bound and soluble forms similar to the interleukin- I receptors (17). These receptors (secreted tumor necrosis factor receptors I and secreted tumor necrosis factor receptors II) regulate the activity of both tumor necrosis factor-a and the less potent form, tumor necrosis factor-ß.

In addition to interleukin-1 and tumor necrosis factor-a, we must consider additional proinflammatory cytokines such as interleukin-6 and related molecules such as the prostaglandins and leukotrienes

and the inhibitors of inflammatory cytokines such as interleukin- 10.

Interleukin-6 Interleukin-6 is produced by macrophages, fibroblasts, lymphocytes and endothelial cells. Produc- tion is induced by interleukin-1 and lipopolysac-

charide and suppressed by estrogen and progesterone. It may be through interleukin-6 that these hormones exert their effects on gingiva. Interleukin- 6 causes fusion of monocytes to form multinuclear cells that resorb bone.

Interleukin-8 Interleukin-8 is produced by a wide variety of cell types, including polymorphonuclear leukocytes,

monocytes, fibroblasts and keratinocytes in response to microorganisms, mitogens and endogenous mediators such as interleukin-1 and tumor necrosis factor. One of the major functions of interleukin-8 is its ability to induce the directional migration of cells, including polymorphonuclear leukocytes,

monocytes and T cells (230), thus playing a key role in the accumulation of leukocytes at sites of inflammation.

Interleukin-10

Interleukin-10 plays a major role in suppressing im-

mune and inflammatory responses. It is produced by T cells, including human ThO, Thi and Th2 cells, B

cells and monocytes and macrophages after activation (55). Interleukin-10 inhibits the antigen-presenting capacity of macrophages by downregulating

class II major histocompatability complex expression. It has also been reported to inhibit antigen presentation by Langerhans cells (77). Human interleukin-10 reduces significantly the proliferation and production of cytokines by both Thi and Th2 clones exposed to specific antigen and phytohemagglutin (55). It also has direct inhibitory effects on inter-

4


feron-y production (54,56). Interleukin-10 has been found to act as a specific chemotactic factor towards CD8+ T cells while suppressing the ability of CD4+ T cells to migrate in response to interleukin-8 (145). Interleukin-10 is a potent growth and differentiation factor for activated human B cells and thus plays an important role in amplifying the humoral immune

response (253). Interleukin-10 inhibits the synthesis of interleukin-la, interleukin-6 and interleukin-8 but enhances the production of the interleukin-1 receptor antagonist, so that this cytokine dampens im-

mune proliferation and the inflammatory response (54,89). Interleukin-10 abrogates antigen presentation by macrophages and indirectly inhibits T-cell

activation (17).

The lymphocyte cytokines The lymphocyte-signaling cytokines interleukin-2, interleukin-3, interleukin-4 and interleukin-5 are all involved in lymphocyte clonal expansion, differen-

tiation of B cells into antibody-producing plasma cells and immunoglobulin isotype switching. In-

terleukin-2 is produced by both CD4+ (helper) and CD8+ (suppresser) (142), activated T lymphocytes

and induces expression of clones of specific T cells and secretion of interleukin-3 and interleukin-4. In-

terleukin-4 regulates immunoglobulin production and suppresses activated macrophages and causes their apoptosis.

Transforming growth factor-ß

Role of matrix metalloproteinases in periodontal disease Loss of attachment, as occurs in periodontal disease, is associated with extensive breakdown of collagen fibers in the periodontal tissues. Since the major structural protein of the periodontium is collagen, the assessment of matrix metalloproteinase levels during inflammation will reflect the association of matrix metalloproteinases in collagen destruction in periodontitis (Fig. 1). The involvement of matrix metalloproteinases in pathological tissue destruction is well documented, and the evidence for their role in periodontal destruction has increased over the years. In earlier studies, collagenases has been identified in explanted gingival tissues and their culture fluids (22,83,249) and in homogenates of gingival biopsies (316). Collagenase activity has also been demonstrated in gingival crevicular fluid, and this activity increased with the severity of inflammation (101). Further evidence that matrix metalloproteinases are involved in tissue destruction in human periodontal disease were summarized by Birkedal- Hansen (22), which includes:

" cells isolated from normal and inflamed gingiva express matrix metalloproteinases in culture;

" several metalloproteinases can be detected in human gingiva cells in vivo;

" polymorphonuclear leukocytes collagenase and gelatinases can be detected in gingival crevicular fluid from gingivitis and periodontitis patients.

Transforming growth factor-ß is produced by activated T cells. It is a chemoattractant for monocytes and it suppresses their activation. It also induces

production of immunoglobulin A and immuno-

globulin G2b. Transforming growth factor-a is produced by macrophages and serves as a mitogen for fibroblasts, epithelial and endothelial cells. Inter- feron-y is produced by CD8+ T cells and is responsible for recruiting and activating macrophages and also upregulates the major histocompatibility

complex on virally infected cells and targets them for killing.

In summary the cytokine network has a crucial role in immune homeostasis. Complex summative, overlapping, synergistic and inhibitory interactions

exist between the various members of this family of molecules. This allows for redundancy within the system and a high level of fine control which prevents devastating effects to the host when dysfunc-

tion or dysregulation occurs.

Each of the major cell types of human periodontal tissues is capable of expressing a unique complement of matrix metalloproteinase when properly stimulated. The cells include polymorphonuclear

Fibroblasts

Collagenase

Bone resorption

Tissue breakdown

°° °o 0

0

Osteoblast Osteoclast Fig. 1. Macrophages produced cytokines (such as interleukin-1) induce fibroblasts and osteoblasts to produce proteases which result in bone and tissue breakdown.

5

Kinane et al.

leukocyte, fibroblast, keratinocyte, macrophage and endothelial cells.

Collagenases in tissues may existed in different

forms, that is, in active form, in inactive form that is an enzyme inhibitor complex, or in latent form

(proenzyme). Total collagenase activity has been

shown to be high in gingival crevicular fluid of inflamed dogs gingiva (161), in gingival crevicular fluid

of localized early-onset periodontitis and chronic adult periodontitis patients compared with gingivitis and control patients (329,330), in gingival crevicular fluid of untreated localized early-onset periodontitis patients and also that the activity decreased after treatment (169). Gingival crevicular fluid collagenase activity has been noted to increase with disease severity, that is, adult periodontitis and localized early-

onset periodontitis > gingivitis > healthy gingiva (329,330). ingman et al. (139) demonstrated that saliva samples did not reflect periodontal tissue de-

struction clearly as they found, only low amount of total collagenase activity in saliva of untreated localized early-onset periodontitis patient and the level

was comparable to normal patients. Ingman et al. (140) also demonstrated that in saliva no difference in collagenolytic activity could be seen in adult periodontitis, localized early-onset periodontitis and control patients.

Latent collagenase levels were found to be higher

in gingival crevicular fluid of gingivitis sites in humans and dogs compared to healthy sites (161,

162,171), and also in normal sites (161). It was also detected in the explant of clinically normal tissue (125). Higher latent collagenase levels seems to reflect either the normal healthy state or the gingivitis state. Little or no active collagenase activity was demonstrated in the gingival crevicular fluid of healthy and gingivitis sites in dogs (167), but higher levels were found in gingival crevicular fluid of localized early-onset periodontitis compared with normal patients (169). Higher levels were also detected in

gingival crevicular fluid of progressive periodontitis patients compared with stable and gingivitis patients (171). In whole saliva, high active collagenase activity was demonstrated in untreated adult periodontitis (85,139) and localized early-onset periodontitis, but as treatment progressed more collagenase is present in the latent form (85). Higher levels of active collagenase correlates with active de-

structive phases of periodontal disease (171). Although there is no direct evidence for a causal

relationship between matrix inetalloprotcinases and periodontal tissue destruction, the evidence for the involvement of collagenase in collagen breakdown

during chronic inflammatory periodontal disease is

strong. Matrix metalloproteinase-8 is detected at high levels in gingival crevicular fluid and saliva during gingivitis or periodontitis, whereas it is undetectable in healthy individuals (139,285). In extracts or homogenates of diseased periodontal tissues, matrix metalloproteinase-1 is abundantly present in contrast to healthy specimens and, as with matrix metalloproteinase-8, a positive correlation was found between its presence and the severity of inflammation (231). In addition, matrix metalloproteinase-1 has been immunolocalized in inflamed but

not in healthy periodontal tissue (348). These data

may suggest that, following production during inflammation, most matrix metalloproteinase-1 remains in the gingival tissue, whereas most released matrix metalloproteinase-8 finds its way to the pocket.

Adhesion molecules It has been recognized that blood vessels play a central role in orchestrating the inflammatory and immune processes. Blood vessels are lined by endothelial cells that are activated by cytokines and bacterial lipopolysaccharide to become more adhesive to circulating polymorphonuclear leuko-

cytes, monocytes, eosinophils, basophils and T and B lymphocytes. Thus, the endothelial lining is transformed from being anti-coagulant to becoming pro-coagulant. The changes in the vessels contribute to the inflammatory changes in periodontal disease, but the most important features is the migration of leukocytes from the circulation. When

the endothelial cells become activated they express adhesion molecules, such as the intercellular adhesion molecule-I, that enable specific subpopulations of leukocytes to bind and subsequently migrate into the tissues under the influence of chemotactic gradients (Fig. 2). Activators include

the proinflammatory cytokines, lipopolysaccharide, interleukin-4 and interferons. In vivo, interleukin-1,

tumor necrosis factor and Iipopolysaccharide will result in rapid accumulation of large numbers of polymorphonuclear leukocytes. In contrast, in-

terleukin-4 and interferon are likely to be more important in the recruitment of a mononuclear cell dominated infiltrate.

Intercellular adhesion molecule- I is expressed on endothelial cells and shows an increase in expression after stimulation with interleukin-I, tumor necrosis factor and interferon (Fig. 2,3). It plays an im-

6


portant part in the adhesion of all leukocytes to endothelial cells and migration. Evidence for this is leukocyte adhesion deficiency, a congenital disease

with low levels of adhesion molecules. Patients sub-

sequently suffer from recurrent infections due to in-

ability of their leukocytes to extravasate into dam-

aged or injured tissue. A constant and characteristic feature of these individuals is a very rapid severe generalized prepubertal periodontitis affecting both

the primary and permanent dentition (335). The endothelial cell leukocyte adhesion molecule-

1 has been found to play an important role in polymorphonuclear leukocytes binding to activated endothelial cells. It is rapidly induced by interleukin-1,

tumor necrosis factor or lipopolysaccharide.

In clinically healthy gingiva, endothelial cell leukocyte adhesion molecule-1 and intercellular adhesion molecule-I are found in postcapillary venules in the vicinity of the gingival crevice, with more dis-

tant vessels being negative. A further finding in the

study of Moughal et al. (211) was the observation

that junctional epithelium expresses intercellular adhesion molecule-1. This specialized epithelium may be adapted to express intercellular adhesion mol-

ecule-1 constitutively even in the absence of interferon-y stimulation. This may reflect the fact that junctional epithelium is the site of extensive poly-

morphonuclear leukocytes migration both in health

and in disease (12,266,269).

Polymorphonuclear leukocytes

Random Rolling Sticking Extravasation contact

E-s Selectins Integrins

4 Leukocyte rolling Blood vessel wall

\ýJy$9ý9piýR� JCAM I

Activation of ' Jý

endothelial LIPS ILl J

cell

Fig. 2. The leucocyte contacts, rolls, sticks and eventually extravasates out of the blood vessel prior to beginning its journey to the site of inflanunation.

Leukocyte rolling 1i Blood vessel wall

oI -ý I cý Ic AI Acuveuon or endothelial 'PS

cell Direct Indirect

LPS y ALSP

ProtryL R TNFe TNFa

vesicle Ln a-

v, PGE2 Protelý-ý y IL-8

cteri -14 `"&

TOOth $UA8C9 Leukocyte MMP

Fig. 3. Molecules, cells and process influencing the increased adherence of leucocytes to bllod vessels so that they can extravasate and begin to chemotact towards the microbes.

The polymorphonuclear leukocyte is a phagocytic

cell and is the predominant leukocyte in blood and in the gingival crevice (12). It forms about 62% of

peripheral white blood cells and about 91% of gingi-

val crevicular cells recovered by aspiration (278). Polymorphonuclear leukocytes are found at sites of both gingival health and disease (12). The polymor-

phonuclear leukocytes is an important cell because,

along with the integrity of the physical surface bar-

rier, it forms the first-line defense against invading

pathogens. Evidence for its crucial role in the defense of the periodontal tissues comes from experiments of nature where neutrophils are depleted or malfunctioning, such as in neutropenia or Chediak-

Higashi syndrome, and there is severe and rapid loss

of alveolar bone and teeth. Polymorphonuclear leukocytes develop from plur-

ipotential bone marrow stem cells and the cells go through several stages prior to maturity. The lifespan

of a polymorphonuclear leukocyte outside the bone

marrow is short, on average of 7 to 58 hours. As polymorphonuclear leukocytes mature they become

more deformable, alter their surface charge and acquire new cell surface receptors (337). Following

these changes mature polymorphonuclear leuko-

cytes can cross endothelial barriers in the bone marrow and enter the bloodstream.

In the circulation polymorphonuclear leukocytes form two pools; circulating and marginated. Mar-

ginated polymorphonuclear leukocytes are stored in the small vessels, particularly of the lung and spleen. They can be mobilized at short notice in

times of acute need. The average range of polymorphonuclear leukocyte numbers in the circulation is 2.5 to 7.5X 109 cells/liter. This number can increase 10 fold within 28 hours in times of acute inflammation. This rapid and substantial increase in polymorphonuclear leukocyte number occurs as a result of several processes; transfer of cells from

the marginated to the circulating pool, increased

7

Kinane ei al.

release of mature cells from the bone marrow and an increase in the rate of maturation of cells within the bone marrow.

Numerous surface receptors are present on polymorphonuclear leukocytes (251) of which there four

major classes:

" receptors for inflammatory mediators and bac-

terial products; " receptors for lymphokines and monokines; " receptors for opsonization (Fc); and " receptors for endothelium and proteins of the

tissue matrix.

These receptors are used for all stages of polymorphonuclear leukocyte emigration from blood vessels, chemotactic movement, recognition of foreign materials, its binding and phagocytosis and for the control of these processes via lymphokines and monokines.

The polymorphonuclear leukocyte has a characteristic light microscopic appearance with a multi- lobed nucleus and numerous cytoplasmic granules. The granular component of the cell consists of many cytoplasmic granules with an outer membrane en- closing the densely packed protein in a mucopolys- accharide matrix.

Three types of granules exist:

" primary or azurophil granules; " secondary or specific granules; and " tertiary or C particles.

Primary granules are identified by their peroxidase content. Positive staining for peroxidase is seen in

the granules, at all stages, and in the secretory apparatus of polymorphonuclear leukocytes during

their formation. The contents of the primary granules are largely responsible for the non-oxidative killing mechanisms of the neutrophil (205), and these

granules contain:

" antimicrobial agents or defensins, which include

myeloperoxidase, lysozyme and cationic proteins; and

" proteinases, which include elastase and cathepsin G.

Secondary granules are peroxidase negative, of low density and are twice as numerous as primary granules in the mature polyniorphonuclear leukocyte.

Their size is variable and they are typically spheres or rods containing:

" antimicrobial agents - lysozyme

" proteinases - collagenase " other molecules

- lactoferrin

- B12 binding protein

- laminin receptor (67 kDa)

- CD11b/CD18 (C3bi)

- fibronectin receptor alpha

- vitronectin receptor alpha.

The release characteristics and kinetics of primary and secondary granules are different. Secondary

granules are released chiefly extracellularly, with the granules being found close to the advancing edge of cells responding to a chemotactic stimulus (349). Evidence of secondary granule release extracellurly is more prominent than that of primary granule release and occurs at an earlier stage. These phenomenon have been interpreted in the past as indicating

a secondary and extracellular function of secondary granules and an intracellular function of primary granules related to phagocytosis (349).

The process of degranulation is calcium depend-

ent and is not related to cell death (301,349) with lysozymal enzymes released without a rise in lactate dehydrogenase, an indicator of cell death. In vitro, granule release can be stimulated by dental plaque (301), binding of immune complexes and complement, binding to nonphagocytosable surfaces and passage through membranes (349).

Polymorphonuclear leukocytes in inflammation

When functioning efficiently against infectious

agents, polymorphonuclear leukocytes leave blood

vessels and move across tissues towards a chemotactic source where they attempt to kill the organism. Adhesion molecules relevant to the movement of polymorphonuclear leukocytes from vessels are expressed on endothelial cells and polymorphonuclear leukocytes (251). During inflammation the

molecules appear before those for other leukocytes,

thus polymorphonuclear leukocytes are the first cells to cross the vessel wall. The immediate adhesion of polymorphonuclear leukocytes is mediated via upregulation of P-selectin on endothelial cells by

thrombin or histamine. Stimulation of post-capillary venules by inflammatory mediators triggers the upregulation of expression of the endothelial cell adhesion molecules (291). This functions over 24

8


hours, increasing the interaction of leukocytes with the cells of the vessel walls. Various endothelial markers including intercellular adhesion molecule- 1, intercellular adhesion molecule-2 and vascular cell adhesion molecule-1 bind to the LeuCAM family

of integrins, which are expressed on polymorphonuclear leukocytes. Leukocyte function-associated

antigen-1 binds specifically to intercellular adhesion molecule-1 and intercellular adhesion molecule-2.

In response to the expression of surface markers, the cells change shape and roll along the blood ves-

sel wall (Fig. 1). They finally stop moving, flatten and their membranes develope a ruffled border. They

then migrate through the endothelial wall. Endo-

thelial cell leukocyte adhesion molecule-1, which is

necessary for this migration, is expressed in response to interleukin-1 and tumor necrosis factor. Interleu- kin-1 and tumor necrosis factor-a thus have a crucial

role in the control of inflammatory mechanisms fol-

lowing the initial stimulus. Polymorphonuclear leukocytes are capable of re-

sponding to many different chemoattractants in-

cluding N-formyl peptides, complement-derived C5a, leukotriene B4, interleukin-8 and platelet-activating factor (124). Chemoattractant receptors span the cell membrane and are GTPase-coupled receptors. Once outside the vessel, cells migrate through the connective tissue towards the chemotactic stimulus (Fig. 4). Directed movement of polymorphonuclear leukocytes comes about by the interac-

tion of specific chemoattractants with surface receptors on its plasma membrane. This binding activates a cascade of secondary intercellular events that lead

to either chemotaxis or the respiratory burst, de-

pendent on the chemotactic agent (199). Neutrophils

stimulated by chemoattractants undergo morphological changes, becoming elongated cells with a ruffled border. During movement, the anterior edge extends forwards and the posterior edge retracts to-

wards the body of the cell. Adherence of the cells to

a substratum occurs via the glycoproteins and proteoglycans of the extracellular matrix. The concentration of extracellular matrix glycoproteins and proteoglycans changes at sites of gingival inflam-

mation (243), and these molecules can become

chemotactic for inflammatory cells. Actual cell movement occurs by the polymerization of globular actin to fibrillar actin, and current models propose that the fibrillar actin is involved in the generation

of the force required for cell movement (229). Chemoattractant receptors exist in two intercon-

vertible forms: high- and low-affinity states (279). Chemotaxis is initiated at receptors by concen-

Plaque

Junctional Epithelium

rf Neutrophil ýýI'L-8+

` E-selectin

0

ö; ̀

Endothelial cells ' v, 0

Clinically Normal Fig. 4. Interleukin-8 is one of the major host produced chemotactic agents to attract neutrophils to the gingival crevice.

trations of chemoattractants that are much lower than those required to produce the respiratory burst. Snyderman & Pike (284) proposed that binding of chemoattractant to high-affinity sites results in

chemotaxis, whereas binding to low-affinity sites results in degranulation. Chemoattractants appear to be most effective in producing chemotaxis in the following order of potency: interleukin-8, leukotriene B4, C5a, N-formyl peptides and platelet-activating factor (199).

Once at the site of the chemotactic stimulus and prior to attachment of the polymorphonuclear leukocyte to the invading organism, opsonization takes place either in the presence or absence of complement and antibody. In the presence of complement and antibody, activated C3 and immunoglob-

ulin G bind their respective polymorphonuclear leukocyte receptors, CR1 and CR3 for complement and Fcy receptor II and Fcy receptor III for antibody. Capsular organisms are able to evade this opsonization and phagocytosis due to the masking of the

antigenic portion of their cell wall by the capsule and by avoiding the deposition of complement on their surface (280).

If antibody or complement are not available, then alternative systems exist for the opsonization of bacteria. Generally, cell wall antigens determine the ease with which an organism is phagocytosed. Lipopoly-

saccharide-binding protein is an acute-phase reactant which binds lipopolysaccharide, the cell wall component of gram-negative bacteria. The binding

9

Kinane et al.

of lipopolysaccharide-binding protein to bacteria

enables binding to the CD14 receptor on polymorphonuclear leukocytes and macrophages, thus enhancing phagocytosis (350). Since gram-negative bacteria are abundant in the gingival crevice and complement and antibody can be locally destroyed (92), it is likely that this system is in operation.

Following opsonization, phagocytosis can occur. Phagocytosis is the process by which the polymorphonuclear leukocyte takes up a particle. It involves

attachment of the phagocyte to the particle and its

subsequent engulfment. It starts with the receptor- ligand binding between polymorphonuclear leuko-

cyte and the microbe. This interaction activates the ingestion phases which involve actin, myosin and actin-binding proteins. The membrane surrounds the particle, pseudopodia are produced and a phagocytic vacuole is formed. Simultaneously cytoplasmic granules fuse with the phagosome membrane to form a phagolysosome,. This final fusion of the membranes and release of granule contents generally results in destruction of the engulfed particle (328).

Once phagocytosed and engulfed by the polymorphonuclear leukocyte, two mechanisms exist to kill

the organism; oxygen- dependent and oxygen-independent mechanisms (323). Oxygen-dependent killing is activated when the polymorphonuclear leuko-

cyte undergoes a respiratory burst coincident with phagocytosis. NADPH in the phagolysosome membrane is activated and reduces 0, to superoxide (340). Following the generation of hydrogen peroxide and OH in this reaction, a second enzyme system involving myeloperoxidase, a primary granule component, becomes active. Myeloperoxidase is released in large quantities but on its own has little effect. In

combination with H20,, myeloperoxidase is capable of catalyzing the oxidation of plasma halides, the most important of which is chloride because of its

concentration. It is at this stage that the interaction between chlorinated oxidants, protease inhibitors

and proteases can occur such that the proteases can act efficiently on their substrate (Fig. 5).

OH is considered the most toxic of all the free radicals. If this reaction occurs in vivo then the rate- limiting step is the availability of iron. Little free iron

is available due to its binding to transferrin and lac-

toferrin extracellularly and apoferritin intracellularly

(259). In the phagocytic vacuole the p11 is reduced

and some iron may be lost from transferrin; however,

lactoferrin retains its ability to hind iron at low pll, thus protecting the cell and limiting the production

of 011 (118).

Oxygen-independent killing is based on the polymorphonuclear leukocyte granule network. Anti- microbial agents that are independent of 0., are released into the phagolysosome during phagocytosis (311). These granule components, including bacteri-

cidal and permeability-increasing protein, chymotrypsin-like cationic protein and defensins (328,340)

and those outlined above, are strongly bacteriocidal

and in most cases effectively kill bacteria (205). In addition to killing bacteria, the proteolytic enzymes, including elastase and the metalloprotefnases, have the potential to damage extracellular tissue and have been suggested to operate in periodontal tissues (22).

It has been demonstrated that sera from localized

early-onset periodontitis patients with decreased

neutrophil function can modify the activity of healthy neutrophils (2). The serum factors responsible for these effects have been identified as being inflammatory mediators such as prostaglandins and cytokines. Small quantities of bacterial lipopolysac-

charide (in the picogram range) can induce the secretion of tumor necrosis factor-a, interleukin- I and prostaglandin E , from monocytes. The production of interleukin- l -tumor necrosis factor-a and tumor necrosis factor-a-prostaglandin E2 are strongly correlated (208). Very small increases in serum cytokine levels can alter neutrophil function. Prostaglandin E,

and cytokines also stimulate fibroblasts to release matrix metalloproteinases, which break down the extracellular matrix. In addition, interleukin-1, tumor necrosis factor-a and prostaglandin E., are potent stimulators of bone resorption. Adherent mononuclear cells from localized early-onset periodontitis patients have been found to secrete higher levels of prostaglandin E and tumor necrosis factor-a than

generalized early-onset periodontitis patients or healthy controls (274). More recently, insulin-de-

pendent diabetic patients with periodontitis have been reported to manifest excessive levels of inflammatory mediators (256-258). It has been suggested that these observations indicate an underlying immune defect of monocytes in certain diagnos-

tic groups (216). The concept of excessive inflammatory mediator

production by monocytes has been proposed as a possible causative pathway for periodontitis, and indeed one of the mechanisms whereby periodontitis can influence other systemic conditions such as cardiovascular disease (216). Early investigations of genetic polytnorphisms of cytokines have attempted to identify a possible link between different periodontal categories and specific alleles related to an

10


1) Inhibition of adhesion or invasion

Oral e0ittum

o

3) Neutralising Antibody

Proteinase 0

Degradation

IgA

CU: OWX)ll

IgG4 Ge: g

2)Complement Activation G1 Cl Alternative Cl Classical

IgG pathway 3 pathway

42 'C3 fIgm

C3 Cb IgG4

ý

gG3 y

C5p-9 Lysis

4) Opsonisation and Phagocytosis

,Q, C3b

b, _ &0

IgG1 gG'

Pha oso

Neutrophil

Fig. 5. The four main processes whereby antibodies can detrimentally affect bacteria: 1) inhibition of adhesion or invasion; 2) complement mediated killing; 3) neutralis-

ability to overproduce cytokines (57,84,102,159). In

addition, a family linkage study of early-onset periodontitis suggested that the COX-1 enzyme system,

which acts as a catalyst for the breakdown of arachidonic acid, one of the products of which is prostaglandin E2, may be associated with susceptibility to

early-onset periodontitis (89). A further host defense defect that may increase

susceptibility to early-onset periodontitis is neutro-

phil chemotactic defects. A high percentage of members of families with a background of localized early-

onset periodontitis have been reported to show ab-

normal neutrophil chemotaxis (323). Analysis of 22 families with localized early-onset periodontitis demonstrated that, in 19 of these families, the pre-

senting localized early-onset periodontitis patient

and their family members all demonstrated neutro-

phil chemotaxis disorders. In these 22 families the

chemotactic defect was found in almost 50% of the

siblings, indicating a dominant mode of inheritance

IgG

Phagoso

0 0

ation of microbial toxins (cf leukotoxin); and 4) opsonisation of microbes prior to phagocytosis and enhancement of phagocytosis.

(323). Whether the trait is due to an intrinsic defect

of neutrophils or is caused by extrinsic factors in the

sera and thus is environmentally influenced remains controversial (2). The neutrophil chemotactic defect is not noted in all populations exhibiting localized

early-onset periodontitis, and a significant number of patients and families with localized early-onset periodontitis do not show evidence of the defect (152,153,232).

Decreased neutrophil chemotaxis in localized

early-onset periodontitis has been linked to a reduced receptors density for chemoattractants such as N-formyl-methyl-leucyl-phenylalanine, complement fragment C5a, leukotriene B4 and interleukin- 8 (49,52). A cell surface glycoprotein (gp110), which is thought to be involved in neutrophil movement and secretion, is reduced in localized early-onset periodontitis patients (324). In addition, several other post-receptor defects have been identified (49). De Nardin (53) found that decreased N-formyl-

11

Kinane et al.

methyl-leucyl-phenylalanine binding was related to

variations in N-formyl-methyl-leucyl-phenylalanine

receptor DNA in localized early-onset periodontitis

patients and healthy controls. These preliminary results indicate a possible role for a N-formyl-methyl- leucyl-phenylalanine receptor alteration in neutrophil chemotaxis. However, many molecules related to neutrophil dysfunction in localized early-onset periodontitis patients belong to the same family of surface receptors and have similar morphogenic characteristics. Similarities in signal transduction

mechanism between cell surface components also exist. De Nardin (53) hypothesized that a common underlying defect, at either the cell surface receptor or post-receptor level, may contribute to the alteration in neutrophil function seen in patients with lo-

calized early-onset periodontitis.

Fc receptors Fc receptors are the receptors found on leukocytes

that can bind immunoglobulin and are thus relevant to the microbial opsonization properties of phagocytes. Some investigators have shown that patients with generalized early-onset periodontitis have high

antibody titers against Actinobacillus actinomycetemcomitans serotype b outer membrane antigens (107,310,342). It has also been suggested that the

subclass IgG2 is a poor opsonin because it fixes complement less effectively than IgG3 and IgG1 and binds weakly to Fc receptors on the surface of phagocytes (250), which tends to complicate theories of protective effects afforded by high anti-A. actinomycetemcomitans titers of IgG2 in black localized

early-onset periodontitis patients. However, a further aspect of the immune response

may yet explain this phenomenon. Fc receptors are the crucial link between the humoral and inflam-

matory components of the host defense system. Fcy

receptors on phagocytes recognize the Fc region of IgG and facilitate opsonization of microbes. It has been shown that neutrophils express low-affinity

variants of Fcy receptors (Fcy receptors II and Fcy

receptors III) under normal conditions. Fcy receptors II binds all subclasses of IgG, but Fcy receptors III only binds IgG3 and IgGI. Many more Fcy receptors III than Fcy receptors II are expressed on the

surface of neutrophils. However, while Fcy receptors III can evoke lysosomal degranulation, they cannot promote the necessary respiratory burst activity and phagocytosis needed for microbial killing of opsonis- ed bacteria. Fcy receptors 11, on the other hand, can

precipitate all three activities. It is possible that Fcy receptors III may prepare polymorphonuclear leukocytes for phagocytosis mediated by Fcy receptors II (343).

It has been reported that the numbers of Fcy receptors II and Fcy receptors III on peripheral blood

neutrophils of localized early-onset periodontitis patients are within the normal range (172). Miyazaki et al. (205) found that both Fcy receptors II and Fcy receptors III expression and neutrophil phagocytosis were significantly depressed in gingival crevicular fluid compared with peripheral blood in adult periodontitis patients. The reduction in Fcy receptors was significantly correlated with phagocytic activity. In addition, the messenger RNA level of Fey receptors III was also significantly lower in gingival crevicular fluid neutrophils than in peripheral blood neutrophils but not the messenger RNA level of Fey receptors II. These findings indicate local suppression of Fcy receptor expression and/or the production of Fcy-binding proteins by periodontal pathogens (130).

Adaptive aspects of the immune response in early-onset periodontal disease

In contrast to the innate immune mechanisms dis-

cussed previously, adaptive immune responses are not initiated at the site where a pathogen first estab- lishes a focus of infection.

The humoral immune response The main function of the humoral immune response is the destruction of extracellular microorganisms and prevention of the spread of intracellular infec-

tions. This is achieved by antibodies secreted by B lymphocytes. Antibodies induced have many func-

tions ranging from neutralization, to leading to opsonization. The activation of B cells and their differ-

entiation into antibody-secreting cells is triggered by

antigen and usually requires helper T cells. Many studies over the years have demonstrated

that antibody responses to various periodontal pathogens are increased in patients with periodontal disease compared with subjects without the disease. Therefore, antibody titers, isotypes and the re-

sponses of the humoral immune system have been

investigated over the years. Mouton et at. (212) established that antibody to

12


Porphyromonas gingivalis is found in a significant proportion of healthy adults as well as being detect-

able in children as young as 6 months. In both adults and children there was a positive correlation be-

tween antibody levels and age. Nevertheless, serum IgG antibody increased by 500-800% in adult periodontitis and generalized early-onset periodontitis patients. Multiple subsequent studies confirmed these findings and showed a significant increases in levels and frequency of antibody to P. gingivalis in

adult periodontitis and a subset of generalized early- onset periodontitis patients (10,11). Chen et al. (39)

noted that only some generalized early-onset periodontitis patients displayed an elevated humoral im-

mune response to P gingivalis, albeit these antibodies were of low avidity and increased following

periodontal therapy. celenligil & Ebersole (32) dem-

onstrated a significantly higher level of antibody to the P gingivalis strains (that is, strains generally de-

rived from United States populations) in United States patients compared with similar disease groups derived from a Turkish population. However, the Turkish early-onset periodontitis patients exhibited a significantly higher frequency of elevated antibody to the P gingivalis compared with a geographically matched normal group. The four concepts originally put forth by Mouton et al. (212) regarding this pathogen were: (i) based on the humoral immune response, P. gingivalis is probably an causative agent in periodontal disease; (ii) the humoral immune response is probably protective; (iii) diseased and healthy individuals can be distinguished in terms of their antibody response to this organism; and (iv) differences in the antibody response characteristics may denote different periodontal disease classifications. To these can now be added that the humoral immune response to P. gingivalis may contribute a better understanding of patient susceptibility, diag-

nostic categories, treatment effects and immune

protection and that this pathogen may have the capacity to express antigenic diversity between subjects in a population and across patient populations.

A wealth of literature supports the existence of lo-

cal specific antibody production by plasma cells present in inflamed tissues of the periodontal pocket (68,72,167,283,295). Likewise, a proportion of gingival crevicular fluid samples within a given subject have local antibody levels significantly greater than

can be accounted for by a serum contribution (68, 71,146). In addition to these inflammatory mediators, evidence has been provided that indicates

that both C3 and C4 components of the complement system are cleaved in gingival crevicular fluid from

early-onset periodontitis (237,265), indicating the potential of antigen-antibody complexes to contribute to local immune modulation. Cross-sectional studies have suggested that those gingival crevicular fluid samples with elevated antibody frequently harbor the homologous bacteria and suggest that a combination of the antigen and the host-response is frequently associated with progressing disease (72); these local antibody levels decreased with pocket depth, attachment level stabilized and inflammation resolved following therapy (72).

Lu et al. (185) have shown that, in certain populations, serum IgG2 levels are increased in localized early-onset periodontitis patients, but not generalized early-onset periodontitis, adult periodontitis or periodontally healthy individuals. IgG2 is the predominant immunoglobulin subclass that reacts with carbohydrate antigens preferentially and is particularly reactive with A. actinomycetemcomitans serotype b outer membrane antigens. A. actinomycetemcomitans is a microbe commonly associated with early-onset periodontitis (see Darby & Curtis in this volume). Serum IgG2 levels increase with age and a rise can be detected at puberty (185). An interesting association is that black subjects in general have higher levels of IgG2 than whites, and the incidence of localized early-onset periodontitis among black individuals has been shown to be up to 15 times greater than among whites (182,201,263). Furthermore, the increase in serum IgG2 levels in localized early-onset periodontitis patients was due to high anti-A. actinomycetemcomitans serotype b levels, which actually only accounted for around 10% of the IgG excess in localized early-onset periodontitis patients. It thus appears that localized early-onset periodontitis patients, particularly black individuals, have a tendency to produce increased levels of IgG2 (310).

A family study of early-onset periodontitis indicated the possibility of genetic transmission of IgG., levels (193). Tew et al. (310) suggest a protective role for high levels of IgG2 in patients with localiazed

early-onset periodontitis. Smoking has been found to suppress the IgG, response in adult periodontitis and generalized early-onset periodontitis patients and is associated with more severe destruction, but this does not appear to occur in localized early-onset periodontitis patients (310). Furthermore, the paradoxically lower IgG2 titers seen in black generalized early-onset periodontitis smokers may be specific to A. actinomycetemcomitans, and the IgG2 response against antigens of other bacteria may not be lower in generalized early-onset periodontitis subjects who smoke (306).

13

Kinane et al.

The cell-mediated immune response

Through nonspecific interactions of the T lympho-

cyte with other cells, controlled by a varying array of adhesion molecules, the naive T-cell migrates through lymph nodes, makes initial interactions with antigen-presenting cells and, after specific activation, eventually moves to peripheral tissues, where it can interact with target cells. This is thought to be

the case during the maturation of T cells in periodontal disease, culminating in the homing of specific T cells to the gingival tissues where they can effect their protective function.

Effector T cells fall into 3 functional classes that detect antigens derived from different types of pathogens, presented by the 2 different classes of major histocompatibility complex molecule. Anti-

gens derived from pathogens that multiply in the

cytosol are carried to the cell surface by major histo-

compatibility complex class I molecules and presented to CD8' cytotoxic T cells, and those derived from extracellular bacteria and toxins are carried to the cell surface by major histocompatibility complex class II molecules and presented to CD4+ T cells. These cells, often referred to as helper cells, can differentiate into two types of effector T-cell; Thl and Th2 cells. Thl cells secrete both interleukin-2 and interferon--i when activated by certain types of T-de-

pendent antigens so that they can enhance cell-mediated responses. The Th2 subset of T helper cells produce interleukin-4, interleukin-5, interleukin-10

and interleukin-13 and thereby promote the hu-

moral immune response. Although much of the literature discusses the im-

portance of the humoral aspect of the immune response in periodontal disease, models used in earlier years suggested that the initial periodontal lesion is

composed mainly of T lymphocytes, with B cells and plasma cells predominating at a later stage (188,272, 273). Studies have been carried out reporting differ-

ences in CD4 : CD8 ratios in lesions and peripheral blood of periodontitis patients (226,308) compared with healthy controls, further indicating that the

cell-mediated immune system is potentially as im-

portant in a discussion of this disease as that of the humoral arm.

It appears that, in terms of periodontal disease,

the antigen is picked up by an antigen-presenting cell, such as a Langerhans cell, at the site of infec-

tion, whereupon it is carried to a primary lymphoid

tissue where the presentation to a circulating naive T

cell takes place. This leads to antigen-specific clonal

activation, where some activated cells become effec-

tor cells and others remain in the circulation as memory cells, naive T cells expressing CD45RA and memory cells CD45RO.

CD45 is a transmembrane tyrosine phosphatase with three variable exons that encode part of its external domain. In naive T cells, high-molecular-weight isoforms CD45RA are found. In memory T cells, the variable exons are removed by alternative splicing of CD45 RNA and this isoform is known as CD45RO. The general dogma indicates that memory T cells (CD45RO`) greatly outnumber naive (CD45RA`) T cells in periodontitis lesions (88,155,168,354). How-

ever, it has also been shown that a proportion of these cells are CD45RA+, which suggests reactivation of the memory cell population by specific but as yet identified antigens in the tissues. It has been reported that memory T cells adhere to vascular endothelial cells and augment their permeability to macromolecules (48). This functional ability of memory T cells to control endothelial permeability and to adhere to endothelial cells maybe a dominant factor that contributes to the preferential migration of memory T cells into

sites of chronic inflammation (240), such as the dis-

eased gingival tissues. The realization that T lymphocytes are present in

large numbers in the diseased tissues of early-onset periodontitis patients and not in health has led to in-

vestigations of the numbers, ratios and functionality

of T cells in both the peripheral blood and diseased

tissues of patients with periodontal disease. Studies of peripheral blood T-lymphocyte subsets in early-onset forms of periodontal disease have revealed a wide variety of contradictory results with either depressed CD4' : CD8 ' ratios (154) or mixed (148), or even not correlated to the disease (76). A number of studies have also looked at the distribution of the T-cell subsets in the tissues. One study reports that the CD4+ : CD8+ ratio in the tissues of early-onset periodontitis patients and even their families tend to be increased (300). A study by Stoufi et al. (295) and one by Celenligil et al. (36), however, clearly states that the

ratio is decreased in patients with this disease. What is obvious and further indicated by a study looking at juvenile periodontitis (202) is that the ratios are highly

variable between different patients with the same dis-

ease, making generalizations and conclusions extremely difficult. What the above study did suggest however, was that the high CD4 ': CD8- ratios were seen in those patients that had a greater degree of inflammatory cell infiltration, while the low CD4' : CD8 * ratio tissues were less inflamed.

When diseased gingival tissue is removed and the

cells extracted, both CD4 ' and CD8' lymphocytes

14


are present in large numbers (296,308). Many

studies have indicated raised levels of CD8 ` cells in

diseased tissues; however, CD4' cells, although

prominent, were found at lower levels than in the

peripheral blood. Overall ratios of CD4' : CD8' were

not found to be altered in early-onset periodontitis

and chronic periodontitis patients (196); however,

these ratios were found to be depressed in patients

with localized early-onset periodontitis and generalized early-onset periodontitis (154).

T-cell functions The role of CD4+ T cells in periodontal disease

T-cell functions in periodontal granulation and gin-

gival tissues can be elucidated by their cytokine synthesis profile (353). There are two main subsets of T- helper cells, Thl and Th2, which are usually deter-

mined by the cytokine profile that is characteristic of them (see Table 1 for characteristics of cytokines that

play an important role in the progression of periodontal disease). ThO cells have also been identified,

and these cells are thought to secrete interleukin-4

and interferon-y. Their actual role is yet to be confirmed; however, it is thought that they may play a role as a precursor cell to "l'-helper cells yet to have

differentiated to become either Thl or Th2 cells (207). Although the cytokine profile is a good tool for

T-cell subset investigations, the results reported are often conflicting, causing confusion. Often the re-

suits cannot assess the relative importance of the Th 1 and Th2 subsets.

Fujihashi et al. (81) investigated ThI and Th2 cytokine messenger RNA expression by CD4' '1' cells from diseased gingival tissues. They detected interferon--y, a Thl cytokine, but failed to detect the Th2

cytokine interleukin-4, thus indicating a more domi-

nant role for Thl cells. Other groups, however, indi-

cate a more dominant role for Th2 cells after detection of interleukin-4, interleukin-5, interleukin-6 (10, 192,355) and interleukin-10 (90). One of the current theories on the Thl/Th2 paradigm is that both cells have an important but different role. Gemmell et at. (89), have suggested that Thl cells remain tightly lo-

calized at sites undergoing an active disease process, whereas Th2 cells are more widely distributed throughout the tissue and typify a more quiescent stage of the disease; that is, the immune response in

periodontal disease is predominantly Th2 driven,

with active phases of the disease leading to a foci of Thl activity.

The role of these cell types is that of help for the immune response. Thl cells, characterized often by their production of interferon-y, are important in

microbial killing and are known to augment cytotoxic T-cell functions through their production of interleukin-2 and interferon-y. Th2 cells, however, inhibit much of the work done by Thl cells. The production of interleukin-4 and interleukin-10 by these cells inhibits the actions of interferon-y, hence has

anti-inflammatory characteristics. The production of

Table 1. The role of cytokines in the pathogenesis of periodontal disease

Cytokine Produced by Role

Interleukin- I Large quantities produced by Proinflammatory properties macrophages Mediator of tissue destruction in peridontal disease (6)

Interleukin-4 Activated T cells Inhibits production of interleukin- 1, tumor necrosis factor and interleukin-6 An absence of interleukin-4 in the periodontal tissues has been

suggested to trigger disease progression (80,132)

Interleukin-6 Lymphoid and non-lymphoid Mediates inflammatory tissue destruction cell types Thought to be an important cytokine in B-cell differentiation and Production is triggered by hence to play a role in the induction of the elevated B-cell response interleukin-1, tumor necrosis in patients with periodontal disease (80) factor and interferon-y

Interleukin-8 Mononuclear monocytes and Attracts and activates neutrophils into the tissues of the many of the tissue cells periodontimn, particularly as it is produced by gingival fibroblasts

(: 304)

Interleukin-12 Monocytes, macrophages, Pleiotrophic effects on natural killer cells and T cells in the tissues 8-cells and accessory cells Induces interferon-y production and is necessary for ThI induction

Tumor necrosis Macrophages and lymphocvtes Similar effects to interleukin- I and interleukin-6 factor a and J1 respectively Interferon y Activated T cells Potent inhibitor of interleukin- 1, tumor necrosis factor-u and tumor

necrosis factor-(3

15

Kinane et al.

interleukin-4, interleukin-5 and interleukin-6 elicits and helps maintain a humoral immune response.

Natural killer cells Natural killer cells are also said to constitute a large

part of the cell-mediated immune system, although they are classed as a component of the innate im-

mune system. These cells are large granular lympho-

cytes that are often detected by the marker CD16, thus making the actual numbers detected very diffi-

cult to report, since this marker is also found on other peripheral blood lymphocytes and monocytes. In contrast to CD8+ cytotoxic T cells and killer cells, which kill infected targets cells in an antigen-specific manner, natural killer cells kill infected target cells or tumor cells in an antigen-nonspecific manner (196).

The role of natural killer cells in periodontal dis-

ease are present is unknown. However, there are reports of an increase in the numbers of natural killer

cells in the peripheral blood of'patients with early- onset periodontitis (35,158,339).

Other studies have reported no increase in the numbers of natural killer cells found in the peripheral blood of diseased patients compared with health (358). In the tissues of diseased patients an increase

of natural killer cells is seen, which is actually not difficult or surprising considering the fact that, in health, very few natural killer cells are present in the gingival tissues. The numbers of natural killer cells in the tissues are seen to increase from health to gingivitis to periodontitis (44,82,351), although the proportion of these cells relative to total lymphocyte

numbers actually decreases (44). The activity of natural killer cells is highly in-

creased by soluble mediators such as interleukin-2, interferon-a, interferon-0 and interferon-y. Inter- ferons a and ß are antiviral proteins synthesized and released by leukocytes, fibroblasts and virally infected cells; interleukin-2 and interferon-y are released by activated T cells (18). Surface lipopolysac-

charide from gram-negative bacteria, however, appear to provide the major activation signal for

natural killer cell-mediated cytotoxicity in periodontal disease (174), which leads to natural killer

cell cytotoxic activity against host cells.

Cytokines in periodontal disease

Interleukin-1

In the periodontal tissues, interleukin-10 predominates over interleukin-la and is present at higher

levels in active diseased sites compared with healthy or stable diseased sites (144,289). The variation in interleukin-1 concentrations between sites in the same individual, and the lack of variation in serum interleukin-1ß levels between adult periodontitis patients and healthy controls, suggests local rather than systemic production of interleukin-1 (38). Acti- vated keratinocytes and Langerhans cells produce interleukin-la and interleukin-1ß (131). Macro- phages, polymorphonuclear leukocytes, B cells and fibroblasts secrete interleukin-1ß (90,144,302). Not surprisingly, interleukin-1 receptor antagonist has been detected in periodontal macrophages and found in the periodontal crevicular fluid of patients with periodontitis (137,147).

Numerous periodontal bacteria can stimulate host cells to produce interleukin-1 (88,90). Levels of interleukin-iß correlate with the periodontal status (242). Interleukin-1 stimulates fibroblasts, endothelial cells, monocytes and granulocytes resident in the gingival connective tissue to express adhesion molecules (303). These adhesion molecules aid the passage of immune response cells from the capillaries into the inflamed tissues.

Interleukin-1 can induce the gingival fibroblast

production of plasminogen activator, interleukin-6

and prostaglandin E2 as well as the matrix metalloproteinases, which are involved in tissue destruction,

turnover and remodeling (246). Both interleukin-1

and tumor necrosis factor-a strongly induce matrix metalloproteinase expression in resident and immi-

grant cells, but interleukin-1 is more potent than tumor necrosis factor-a (22,318). Interleukin-1 stimulates osteoclasts to resorb bone (136) while inhibiting the formation of bone (288). More specifically, interleukin-1-mediated resorption of bone can be induced by periodontal pathogens (141).

Tumor necrosis factor

The proinflammatory cytokines tumor necrosis fac-

tor-a, interleukin- la and interleukin-1 ß have been identified in periodontitis lesions from patients with adult periodontitis at higher levels than in healthy

sites (197,290). However, cells containing interleukin-1 were found in much greater numbers than those producing tumor necrosis factor-a and in-

terleukin-la (290). In contrast to interleukin-1, only very low concentrations of tumor necrosis factor-a have been demonstrated in gingival crevicular fluid from both adult and early-onset periodontitis patients (252,357). However, high levels of tumor necrosis factor-a in periodontal lesions appear to rep-

16


resent an established inflammatory and immune response. Salvi et al. (256) examined gingival crevicular fluid interleukin-1 ß, tumor necrosis factor-a and prostaglandin E2 levels in insulin-dependent diabetes mellitus patients with periodontal disease. They found that diabetics had significantly increased levels of prostaglandin E2 and interleukin-1 ß but not tumor necrosis factor-a, compared with non-diabetic controls with similar periodontal status. In addition, diabetics with moderate to severe disease had

almost two-fold higher levels of prostaglandin E2

and interleukin-1 p compared with diabetics with gingivitis or mild periodontitis.

Prostaglandin E2 causes increased vascular dilation and permeability. It also stimulates macrophages to secrete matrix metalloproteinases and has been shown to trigger bone resorption in vitro (22),

thus acting synergistically with interleukin- 1 and tu-

mor necrosis factor-a. Furthermore, it upregulates receptors for complement and immunoglobulin on monocytes and neutrophils (216). The major source of prostaglandins and leukotrienes in inflamed periodontal tissue is the activated macrophages, although they can also be produced by fibroblasts. Prostaglandins, especially prostaglandin Ems, comprise the primary pathway of alveolar bone destruc-

tion in periodontitis. Leukotrienes, especially leuko-

triene B4, are potent chemoattractants for polymorphonuclear leukocytes.

odontitis) than from healthy or gingivitis sections. They suggested that in gingivitis interleukin-10 might suppress inflammation by decreasing macrophage activity, thereby preventing progression to periodontitis. No differences were found in the percentages of interleukin-10* CD4+ cells between the two disease categories. However, some individuals were found to have higher numbers of interleukin- 10+ T cells regardless of disease category. It appears that the overall variation between the two diseases entities in interleukin-10+ CD8+ cells may have been due to individual variation in interleukin-10 secretion patterns rather than differences in pathogenesis between the two disease categories. Further studies are needed to elucidate the role of interleukin-10 in periodontal disease.

In summary, interleukin-1, interleukin-6 and tumor necrosis factor are proinflammatory cytokines which are detected and may be modulated within the periodontium. They are important and ubiquitous aspects of the host defense system, operating in the periodontal tissues during health, destruction and healing phases. Similar claims may in future be made for other members of the cytokine network, in particular interleukin-10 and various soluble and cellbound receptors and inhibitors of cytokine function. These molecules are inter-related and are part of the immune, inflammatory, breakdown and repair homeostasis ongoing within the periodontium.

Interleukin-10

Interleukin-10 may have a role in the progression of adult periodontitis. This was suggested by Gemmell

et al. (90), who found that T-cell lines from both the

peripheral blood of these patients and their gingival tissue secreted interleukin-10, but clones derived from the peripheral blood of an individual with gingivitis did not. It has also been suggested by Stein et al. (292,293) that interleukin-10 may contribute to

autoimmune reactions against gingival tissues. In-

terleukin-10 is able to favor the development of a particular clone of B cells (CD5), which can secrete high levels of autoantibody and is found in increased

numbers in inflamed gingival sections (297). It was found that a subset of type I diabetics may be more susceptible to developing periodontitis through an autoimmune type of reaction directed against gingival connective tissue when exposed to periodontal pathogens (292,293).

Gemmell & Seymour (90) have demonstrated a significantly reduced number of interleukin-10' CD8' cells from periodontitis lesions (adult peri-

The pathology of the periodontal diseases affecting children and adolescents Gingivitis

Various hypotheses have been made over the years linking the severity of gingivitis in childhood with various factors including puberty.

An increase in the sex steroid hormones during the period of puberty is believed to have a transient effect on the inflammatory status of the gingiva (194). Enlargement of the gingivae occurs in both

males and females in areas of local irritation. The inflammation has been described as marginal in distribution, characterized by prominent bulbous

proximal papillae. Enlargement is often only found

on the facial gingiva, whereas the lingual surfaces remain unaltered (98). Although this view is widely accepted, the evidence obtained from cross-sectional and longitudinal studies for this hypothesis is

mainly circumstantial and often fragmented, although still of importance.

17

Kinane et al.

Sutcliffe (299) reported a 6-year longitudinal study that had been carried out to observe the changes in

gingivitis in 127 children 11-17 years old, and the

relationship of these changes with oral hygiene. He

reported an increase in gingivitis associated with puberty. It was noted that girls tended to experience their maximum gingivitis before boys, girls at a mean age of 12 years and 10 months and boys at 13 years and 7 months. Hence, he noted that the distributions

of age were consistent with the hypothesis that there is an increase in gingivitis associated with puberty.

The problems with this and further studies is that they only provide circumstantial evidence linking

puberty and gingivitis. Thus, the association be-

tween the increase in sex hormone levels and gingivitis needs confirmation. Chronological age is a poor indicator of puberty, the measurement of the maximum gingivitis experience is extremely subjective and, further, none of the studies reported have

measured the hormone levels of the subjects. An

alternative explanation for increased gingivitis during puberty is that this is a period of mixed dentition

where erupting and exfoliating teeth present many sites for plaque retention. The fact that gingivitis de-

creases after puberty may reflect the general im-

provement seen in children as they improve their dexterity and become more aware of oral hygiene.

Other clinical indications of a hormonal effect in-

clude fluctuations in gingivitis with phases of the

menstrual cycle (133,176), increased incidence and severity of gingivitis during pregnancy (27,138), gingival enlargement induced by oral contraceptives (149, 187,287) and circumstantial evidence suggesting that there is an increased number of desquamative gingival lesions in menopausal females (216). The mechanisms by which the hormones may influence the

gingiva have so far not been fully elucidated. Steroid hormone receptors are located in hor-

mone-sensitive tissues, where the hormones tend to be retained and allowed to accumulate. In a number of species accumulation of androgens, estrogens and progestins have been observed in the periodontium (194), and particularly pertinent are auto radiographic studies that have demonstrated nuclear localization

of estradiol in human gingival epithelium and gingival fibroblast cells (332-334). The actual accumulation and effect of sex steroid hormone on the gingiva is unknown; however, a number of theories have arisen.

Microbial effects of hormones

Microbial plaque and its presence is a very import-

ant factor on the onset, progression and severity of

gingivitis. It has been reported in other studies of a hormonal nature that microorganisms are effected. Kornman & Loesche (138,149) reported higher levels

of P intermedia in pregnant females with elevated levels of sex hormones and also in women taking oral contraceptives. Therefore, it seems logical that if levels of sex hormones can effect the microflora at one time they must be affected during other periods of hormonal fluctuation. Many studies have been

carried out often leading to reports of speculative results. Several reports of a transient increase of black

pigmented gram-negative anaerobic rods in children during puberty have arisen (46,47,347). However,

none of these studies have used hormone levels as a measure of puberty, and therefore it is difficult to draw inference in this situation.

Vasculature effects of hormones

Investigations in animals suggest that the sex hor-

mones, particularly progesterone, may produce in-

creased inflammation by generating changes in the gingival vasculature (138,149). Local or systemic administration of progesterone produces an increase in

vascularity in ear chamber wounds in rabbits (178). Following systemic administration of progesterone, certain observations suggested that the hormone

might induce alterations in the plasma-endothelial lining of the post-capillary venules (178), resulting in

an increased leakage of plasma proteins and leuco-

cytes. Gingival exudate also increased following administration of sex hormones to dogs (177), and progesterone enhanced acute inflammation during

wound healing in rabbits (220).

Immune effects of hormones

The importance of sex hormones is emphasized by

the findings related to gender-related disease susceptibility. For example, a number of diseases can be

modulated by hormones, mainly estrogens, such as rheumatoid arthritis (128) and Graves disease (8),

which are both affected during pregnancy. Therefore,

some immune responses and reactions are affected in the gingiva leading to a possible role for sex hor-

mones in this disease. The degree of influence of this is unknown, however.

Cellular effects of hormones

Sex hormones are known to effect the cells of the body. Histological studies have shown that estrogens increase epithelial keratinization, stimulate prolifer-

18


ation (247,248,362) and increase the downgrowth

of epithelial attachment (219). Androgens have also been shown to have effects.

Human gingiva is capable of metabolising sex hor-

mones (334), and receptors for estrogen have been found in gingival tissues (16). Furthermore, inflamed

gingiva has been found to metabolize progesterone more rapidly, and these metabolites differ from

those produced by healthy gingiva (121). Vittek et al. (334) have shown a positive correlation between progesterone and gingival inflammation. It is known

that the gingival levels of progesterone and its metabolites increase during pregnancy and that these

metabolites increase the inflammation in pre- existing gingivitis (225). These effects, produced by

progesterone alone or in combination with estrogen, could account for many of the changes seen in puberty and other conditions in which there are in-

creased circulating sex hormones. In the treatment

of children undergoing puberty with gingivitis, the increased susceptibility to plaque-induced inflam-

matory changes should be reduced by the use of optimal levels of plaque control.

The pathogenesis and host responses of gingivitis and periodontitis The normal healthy gingiva is characterized by its

pink color and its firm consistency. Interdentally, the healthy gingival tissues are firm and do not bleed on gentle probing and fill the space below the contact areas between the teeth. Healthy gingiva often exhibits a stippled appearance and there is a knife-

edge margin between the soft tissue and the tooth. In theory, normal gingiva are free from histological

evidence of inflammation, but this ideal condition is

very rarely seen due to the ever present microbial plaque. Even in the very healthy state, the gingiva has a leukocyte infiltrate that is predominantly comprised of neutrophils or polymorphonuclear leuko-

cytes. The primary purpose of these phagocytes is to kill bacteria, which they do by emigrating through

and out of the tissues into the gingival crevicular

area. Clinical changes in the early stages of inflam-

mation of the gingival tissues following plaque accumulation are subtle, but quite rapidly the gingiva

appears red, swollen and the soft tissue has an in-

creased tendency to bleed on gentle probing. Histo- logically, this stage is described by increased vascular dilation and permeability, which leads to an influx

of exudative fluid, proteins and inflammatory cells into the tissues giving them their swollen appear-

ante. At this stage there is intense recruitment of polymorphonuclear leukocytes, macrophages and their progenitor cells monocytes into the tissues. These leukocytes migrate up a chemoattractant gradient to the crevice and are aided by adhesion molecules such as intercellular adhesion molecule-1 and endothelial cell leukocyte adhesion moelcule-1, which assist leukocyte binding to the post-capillary venules and help cells to leave the vessels (151). In addition, as the microbes damage the epithelial cells, the epithelial cells release cytokines, which further encourage recruitment of leukocytes (predominantly neutrophils) into the crevice. The neutrophils within the crevice can phagocytose and digest bacteria and so remove these bacteria from the pocket. In addition to this, if the neutrophil is overloaded with bacteria, it degranulates or explodes and, by doing

so, causes tissue damage due to the toxic enzymes it releases. The neutrophil can thus be viewed as being both helpful and potentially harmful. This neutrophil defense may in some instances operate well and reduce the bacterial load and can be considered important in preventing the gingivitis lesion from being

established. If, however, there is an overload of microbial plaque, then the neutrophils and the barrier that is comprised of epithelial cells will not be sufficient.

After approximately 7 days of plaque accumulation, the initial lesion is said to progress to the early lesion (26,238,267,268). Lymphocytes and macrophages predominate, and the infiltrate occupies 10-15% of the gingival connective tissue. Both

clinically and histologically the changes are very visible. The established lesion as defined by Page & Schroeder (233) is one dominated by plasma cells. Plasma cells have been shown to be predominantly IgG-producing cells, with a lower proportion being IgA (188,189,223,336). In early-onset periodontitis, local IgG4-producing cells in particular seem to be

elevated. Clinically in the established lesion, collagen loss continues both laterally and apically and the dentogingival epithelium continues to proliferate. Two types of established lesion, however, appear to exist. One remains stable not progressing for months or years (175,234), and the second becomes more active and converts to progressive destructive lesions. The very nature of early-onset periodontitis and the knowledge that it is a rapidly progressing destructive disease make it clear that the established lesion here will progress speedily to the destructive

advanced lesion. The advanced lesion is characterized by hone loss, periodontal ligament loss, apical migration of the junctional epithelium to form a true

19

Kinane et al.

pocket and histopathologically by an increase to over 50% of plasma cells in the infiltrated connective tissues (151).

Gingivitis is primarily a response to the bacteria in plaque. It includes a vascular response with in-

creased fluid accumulation and inflammatory cell infiltration. Serum IgG antibody levels to Actinomyc-

es spp. were found to be higher in gingivitis (73), and a similar study showed that antibodies to at least

three bacteria (Actinomyces spp., Bacterionema ma- truchotti and Leptotrichia buccalis) were detected in

gingivitis patients with a wide variation in levels (73). Multiple reports have suggested that gingivitis patients can present with antibody to suspected oral pathogens (50,228), including P. gingivalis and A.

actinomycetemcomitans. Bimstein & Ebersole (21)

noted that IgM antibody levels were significantly different to many bacteria when comparing children versus adults with gingivitis. In contrast, the IgG levels were unrelated to disease in the adult groups but did help to distinguish between the children with and without gingivitis. Thus, generally, the gingivitis patients exhibit antibody to a wide array of these bacteria; however, the levels to suspected periodontal pathogens are uniformly significantly lower

than those seen in periodontitis patients (62,63). Fi-

nally, we observed lower IgA levels in gingivitis sites compared to healthy sites of normal subjects (65),

and Grbic et al. (103) demonstrated that IgA levels

were significantly increased in gingival crevicular fluid from gingivitis sites compared with periodontitis sites, suggesting the potential for IgA to function as a protective factor in this local environment. Generally low levels of antibody to bacteria

associated with periodontitis are found in gingival crevicular fluid from gingivitis sites.

The progression from gingivitis to periodontitis Brecx et al. (25) suggests that more than 6 months may be needed in the normal situation for gingivitis to change to a periodontitis lesion. Furthermore, this move from gingivitis to periodontitis probably only occurs in 10% to 150 of the population. The reason why some patients develop periodontitis more readily than others is highly elusive and thought to be

multifactorial. It should be borne in mind that the

changes during gingivitis, even in prolonged established gingivitis lesion, are to a large extent reversible, whereas the changes during periodontitis, that is, the bone loss and apical migration of the epithelial attachment, are irreversible. Periodontitis is a

cumulative condition whereby once there is bone loss, it is almost impossible to have it re-established and therefore, patients lose bone incrementally over the years. The worst effects of periodontitis are therefore, commonly seen in older patients rather than younger patients, who have still to develop these deep pockets or this extensive bone loss or the associated gingival recession that occurs in periodontitis. The early-onset forms of periodontitis are, however, situations where the normal slow destructive process is greatly accelerated.

Differences between chronic gingivitis and periodontitis The similarities in the inflammatory infiltrate between the stable established chronic gingivitis lesion and advanced periodontitis lesions have encouraged many investigators to look for qualitative and quantitative differences that may be involved in deter-

mining the progression of gingivitis to destructive

periodontitis. Seymour et al. (271) hypothesized that a change

from T-cell to B-cell dominance causes periodontitis. However, Page (236) has disagreed with this view, as have Gillett et al. (97), who showed a predominantly B-cell infiltrate to be associated with stable, non- progressive lesions in childhood gingivitis. Liljenberg et al. (173) compared plasma cell densities in sites with active progressive periodontitis and in sites with deep pockets and gingivitis but no significant attachment loss over a 2-year period. The density of plasma cells (51.3%) was very much increased in active sites as compared with inactive sites (31.0%). It is now generally accepted that plasma cells are the dominant cell type in the advanced lesion (86).

Necrotizing ulcerative gingivitis and periodontitis Necrotizing ulcerative gingivitis is an acute inflam-

matory condition associated with a fusospirochaetal

microbial complex. Necrotizing ulcerative gingivitis begins as linear erythema of the marginal gingiva and extends to form ulcerated, punched-out, painful and necrotic papillae and later may extend to ulcer- ation of the gingival margins (134). The ulcers are covered by a grayish slough, which if removed causes bleeding from the exposed ulcers. The condition is

painful and is associated with marked halitosis often termed foetor oris. As the necrosis progresses, deep bone cratering appears in the interproximal regions in the areas where the papillae have been lost. If left

20


untreated, necrotizing ulcerative gingivitis will progress to necrotising ulcerative periodontitis, which involves destruction of the interproximal alveolar bone and the periodontal ligament (134,198). Marked gingival recession and sequestra formation is a sequel of the rapid necrosis. In industrialized

countries, adolescents and young adults are most predisposed to this condition, and a prevalence of approximately 0.5% has been reported (135).

In a classic electron microscopic study by Listgart-

en in 1965 (179) four zones were identified. These

zones are the: 1) bacteria-rich zone or plaque zone, 2) neutrophil-rich zone in which neutrophils predominate; 3) necrotic zone, which has dead cell, spirochetes and other bacteria; and 4) the spirochet- al infiltration zone, in which tissue structures is pre- served but many invading spirochetes are noted.

Necrotizing ulcerative gingivitis has been characterized by the emergence of selected members of the oral microbiota: Treponema spp. and P. intermedia, which develop as a response to altered resistance to infection of the host tissue (immunosup-

pression). These oral spirochetes have been shown to be capable of invading the infected host tissue (79,181,255). IgG antibody to oral spirochetes (Tre-

ponema denticola and Treponema vincentii) has been reported; the data have shown below normal to selected elevations in disease subjects (14,41,143, 166,213,270,346). Therefore, while treponemal species are frequent in subgingival plaque and comprise a major proportion of some plaques in periodontitis and necrotizing ulcerative gingivitis (281,282), there does not appear to be an elevation in systemic antibody responses, indicative of an active host response, to the cultured treponemal species. Serum

antibody to Actinomyces viscosus, Actinomyces israe- lii, Prevotella melaninogenica and P. gingivalis were within normal limits in necrotizing ulcerative gingivitis during acute and convalescence; however, antibody to Prevotella intermedia was significantly elevated in convalescent sera (73), suggesting that acute disease may exhibit a specific pathogenic microbiota to which the host responds. The contribution of this

response to resolution of the disease remains unknown.

Cancrum oris or noma is a devastating orofacial gangrenous disease that is considered a very severe form of necrotizing ulcerative periodontitis and is

commonly reported in underprivileged African

children (78). The children often suffer from chronic malnutrition, numerous endemic communicable diseases, including viral diseases, and severe adverse physical conditions. These severe debilitating factors

may deplete the subject's immune and inflammatory

system. Measles is the most common infection preceding the development of noma in Nigerian children (78). Acquired immunodeficiency in combination with the impaired endocrine balance in malnourished children results in widespread infection with the measles virus. Anergy resulting from the combination of malnutrition and measles virus infection may promote overgrowth and invasion of anaerobic organisms, including gram-negative bac- cilli and spirochetes. Due to the severe debilitation

of the malnourished child, the infection is not confined locally as necrotizing ulcerative gingivitis but

spreads rapidly across normal anatomical barriers. Severe necrosis and possible sequestration as exemplified by noma are then seen. Minimal data are available on host responses in this disease.

Murray et al. (214) also demonstrated that the microbiota of periodontitis in human immunodeficiency virus (HIV) patients was similar to that seen in

systemically healthy periodontitis individuals; how-

ever, there were significant increases in the numbers of Campylobacter spp. and Treponema spp. Serum IgG antibody to P. intermedia, P. gingivalis and Fuso- bacterium nucleatum have been found to be elevated in HIV-positive homosexuals compared with control normals and seronegative homosexuals (105). Both necrotizing ulcerative gingivitis and necrotizing ulcerative periodontitis (74) occur in a proportion of HIV-infected patients. Rams et al. (244)

compared the distribution of several bacterial morphotypes in necrotizing ulcerative periodontitis and clinically normal periodontitis subjects and found that the morphotypes were very similar from areas of necrotizing ulcerative gingivitis-like tissue necrotic lesions and necrotizing ulcerative periodontitis sites. Importantly, all the patients with the necrotizing ulcerative gingivitis-like lesions had significantly high levels of spirochetes in their pockets. However, the levels of serum antibody to oral microorganisms in HIV subjects has had minimal investigation. Grieve

et al. (104) suggested that HIV periodontitis (necrotizing ulcerative periodontitis) patients exhibited elevations in serum antibody to P. gingivalis, P. inter-

media, F. nucleatum, A. actinomycetemcomitans and Eikenella corrodens. We recently examined serum IgG antibody to a battery of oral pathogens, including T denticola, in HIV patients. We found that,

while some of the patients with necrotizing ulcerative periodontitis exhibited elevated antibody to

suspected periodontal pathogens, the antibody level

to T. denticola isolates was low, and selected HIV patients had minimal antibody to all of the oral micro-

21

Kinane et al.

organisms. Additional studies will be required to delineate the capability of the host response in manag- ing periodontal infections in these patients.

Pathogenic features of adult and early-onset periodontitis Experimental short-term clinical studies have shown that microorganisms quickly colonize clean tooth surfaces when an individual stops toothbrushing. Within a few days, microscopical and clinical signs of gingivitis are then apparent. These inflammatory

changes can be resolved when adequate tooth- cleaning methods are resumed (183,317). Thus,

microorganisms that form dental plaque and cause gingivitis probably do so by the release of components from the bacterium that induce inflammation in the tissues. Short- and long-term clinical trials have underlined the importance of supragingival plaque and subgingival microbial plaque in both the treatment of gingivitis and petiodontitis. Further-

more, animal experiments have indicated that gingivitis only develops in animals that accumulate bac-

terial plaque deposits. Gingivitis is necessary prior to the tissue developing periodontitis, and this implies

that prevention of gingivitis is also a primary preven- tive measure for periodontitis.

The pathogenic processes are largely the response to microbe-induced destructive processes. These

processes are initiated by the microbes but are undertaken by the host cells, and thus it is the host

tissue itself that results in the destruction seen. The host produces enzymes that break down host tissue. This is a necessary process that the host initiates and controls in order to allow the tissues to retreat from

the microbial destructive lesions. Microbial plaque accumulation occurs on the sur-

face of the teeth adjacent to gingival tissues, bringing

the oral sulcular and junctional epithelium into contact with colonizing bacteria and their waste products such as proteases, enzymes, lipopolysaccharide

and toxins. As these products build up, the irritation

to the tissues increases, until production of chemical mediators of inflammation commences and a host inflammatory response is induced.

Elements of the host response that could differentiate adult and early-onset forms of periodontitis

The various aspects of the host response that may contribute to peridontal disease have been covered generally, but several aspects supported within the

literature may be specifically relevant to early-onset periodontitis. These include aspects of the innate, inflammatory and immune defense systems.

The innate defense system includes epithelial, connective tissue elements and soluble products that are antimicrobial and in the normal healthy state may protect against periodontal disease. There are, however, defects reported in epithelial, connective tissue (fibroblasts and cementum), body fluids (saliva) and enzymes (alkaline phosphatase) that may be relevant to early-onset periodontitis (5,9,15, 45). Structural defects such as the absence of endothelial adressins for leukocyte adherence and subsequent migration from the bloodstream into lesions (seen in leukocyte adhesion deficiency with marked periodontal destruction) and defects of epithelium and connective tissue (as seen in Papillon-Lefevre syndrome) as well as more obvious periodontal defects such as that of hypophosphatasia (where defective cementum results in early tooth loss) can all have major effects on the periodontal tissues and lead to early loss of teeth.

Pathogens

The microbes involved in adult periodontitis are largely gram-negative anaerobic bacilli with some anaerobic cocci and a large quantity of anaerobic spirochaetes. The main organisms linked with deep destructive periodontal lesions are P. gingivalis, P. intermedia, Bacteroides forsythus, A. actinomycetemcomitans and T. denticola (360).

P. gingivalis is more frequently detected in severe adult periodontitis, in destructive forms of disease

and in active lesions than in health or gingivitis or edentulous subjects. P gingivalis is reduced in numbers in successfully treated sites but is seen in sites with recurrence of disease after therapy (16,112, 326). It has been shown that P. gingivalis induces elevated systemic and local antibody responses in

subjects with various forms of periodontitis (61). P. intermedia levels have been shown to be elevated in certain forms of periodontitis (9). Elevated serum antibodies to this species have been observed in

some but not all subjects with refractory periodontitis (31). B. forsythus has been found in higher

numbers in sites exhibiting destructive periodontal disease than in gingivitis or healthy sites (47). In addition, B. forsythus was detected more frequently in

active periodontal lesions (13). Although spirochetes such as T. denticola are notoriously difficult to culture in the laboratory, these strict anaerobes may comprise more than 30% of the periodontal subgin-

22


gival microfiora. The above mentioned putative pathogens are always part of a large and varied microflora found in the subgingival plaque. Some are not detected in certain sites with periodontitis and can even be completely absent in cultures from

multiple sites within an untreated periodontitis patient.

The dominant microorganisms involved in early- onset periodontitis include A. actinomycetemcomitans, Capnocytophaga spp., Eikenella spp., P intermedia and anaerobic rods such as Campylobacter rectus (312). Much of the literature regarding the bacterial

effects in early-onset periodontitis have centred around studies on A. actinomycetemcomitans. A.

actinomycetemcomitans is a short, facultatively anaerobic gram-negative rod that is nonmotile and regarded as a key microorganism in early-onset periodontitis. Studies of association have shown in-

creased levels of A. actinomycetemcomitans and increased frequency of antibody to A. actinomycetemcomitans in localized early onset periodontitis (24). A later study by these researchers (25) showed a significantly increased level of IgG antibody to A.

actinomycetemcomitans serotype b in 90% of localized early-onset periodontitis patients, 40% of generalized early-onset periodontitis and only 25% of adult periodontitis patients. Furthermore, clinical studies carried out to observe the effect of treatment

on A. actinomycetemcomitans load have related poor therapeutic outcome with an inability to reduce the

subgingival load of A. actinomycetemcomitans (113, 114,214).

In localized early-onset periodontitis when present, A. actinomycetemcomitans, is predominant. Genco et al. (91) indicated that organisms of the normal flora play a key role in gingivitis, while exogenous organisms or more unusual anaerobic organisms seem to be implicated in periodontitis. Im-

munological studies involving patients with localised

early-onset periodontitis indicate significantly elevated titers of serum antibodies to A. actinomycetemcomitans in localized early-onset periodontitis patients (67,69,73,95,180,191,260,315,331). These

patients have also been seen to produce antibodies locally to A. actinomycetemcomitans (260).

Accompanying the documentation of specific infections in periodontitis patients has been the defi-

nition of active host systemic immune responses to these bacteria (73). In particular, it appears clear that, in most populations, patients with localized juvenile periodontitis exhibit elevated serum IgG

antibody to A. actinomycetemcomitans (30,73). These results have been derived from cross-sectional

studies and demonstrated some correlation with the ability to identify the homologous microorganism in the subgingival plaque (66,309). Accumulating evidence has supported the concept that the predominant serotype of A. actinomycetemcomitans appears to differ throughout the world (11,40,60,75,99,100, 109,210,224,262,314,352,359) and may represent some more general genotypic, phenotypic, and pathogenetic heterogeneity in this species. Ebersole et al. (30) reported a study that examined the frequency of oral disease in an adolescent population and assessed the relationship to A. actinomycetemcomitans. Heavy accumulations of plaque and calcu- lus were frequently observed and were associated with gingival inflammation and, significantly, 25.7% of the students exhibited early-onset periodontitis with 1.7% diagnosed as localized early-onset periodontitis. The findings supported the hypothesis that A. actinomycetemcomitans may represent a pathogen in periodontitis and, while oral health care may be poor, contact with the microorganism appears to be required to initiate disease in this population (30). Ebersole et al. have reported humoral responses in periodontitis patients in various Turkish populations (32). All localized early-onset periodontitis patients from Turkey exhibited elevated antibody levels to A. actinomycetemcomitans, particularly serotypes c and a, while antibody levels to A. actinomycetemcomitans serotype b were highest in United States localized early-onset periodontitis patients. Fifty percent of the Turkish generalized early-onset periodontitis patients also showed elevated anti-A. actinomycetemcomitans antibody. These data suggested that considerable variation exists in the systemic antibody levels to periodontal pathogens and support potential differences in subgingival colonization or antigenic composition of these pathogens between patient populations from differ-

ent geographical regions. Additional studies by this group have also noted elevated serum antibody responses to A. actinomycetemcomitans and selected other oral periodontopathogens in Sjögren's syndrome patients with periodontitis (34) and in patients with Behcet's disease and periodontitis (33). Thus, numerous investigations of many ethnic and geographically disparate groups tend to confirm that A. actinomycetemcomitans is frequently associated with localized early-onset periodontitis, and probably an causative agent in the vast majority of cases and in a substantial proportion of other early-onset periodontitis patients.

Our findings and those of others (23,28,235,277, 338,344,345) suggest that human serum antibody

23

Kinane et al.

reactivities to A. actinomycetemcomitans are observed to a wide variety of antigens derived from the outer membrane of this pathogen. The results indi-

cated a response to certain antigens that were dis-

tinctive in active disease patients and may: (i) indi-

cate changes in antigen expression by A. actinomycetemcomitans (115); (ii) reflect the overall increase in

serum antibody levels noted in active disease patients; or (iii) relate to the increase in the A. actinomycetemcomitans burden, resulting in greater antigenic load, (iv) reflect the extent of existing disease (59) or (v) be associated with more severe disease.

The previously mentioned studies concentrated their efforts on analysis of serum IgG antibody to A.

actinomycetemcomitans or antigens isolated from

the bacteria. The IgG isotype of immunoglobulin

consists of four subclasses, IgGl-4, which have been

shown to be elicited selectively by certain types of antigens and to have diverse functions. Thus, the isotype and subclass of antibodies in the patient serum specific to A. actinomycettmcomitans may reflect restrictions on the capacity of host antibody to bind to particular antigens (61,63,64,96,157). The

repertoire of immunoglobulin isotypes and IgG subclass responses to particular antigens may account for variations in host resistance to infection and dis-

ease. A wealth of literature supports the existence of lo-

cal specific antibody production by plasma cells present in inflamed tissues of the periodontal pocket (68,72,167,283,294). Likewise, a proportion of gingival crevicular fluid samples within a given subject have local antibody levels significantly greater than can be accounted for by a serum contribution (68, 71,72,146). In addition to these inflammatory mediators, evidence has been provided that indicates

that both C3 and C4 components of the complement system are cleaved in gingival crevicular fluid from

early-onset periodontitis (237,265), indicating the potential of antigen-antibody complexes to contribute to local immune modulation. Cross-sectional

studies have suggested that the gingival crevicular fluid samples with elevated antibody frequently harbor the homologous bacteria and suggest that a combination of the antigen and the host-response is frequently associated with progressing disease (72). These local antibody levels decreased with pocket depth, attachment level stabilized and inflammation

resolved following therapy (72). Studies on early-onset periodontitis patients dem-

onstrating infection with A. actinomycetemcomitans have demonstrated elevated gingival crevicular fluid

antibodies to A. actinomycetemcornitans in the ma-

jority of the patients (116,117,260). The presence of all subclasses of IgG have been identified in gingival crevicular fluid, with IgGi, and/or IgG4 levels in gingival crevicular fluid were elevated relative to serum concentrations (241,245). The results support a unique local response in individual sites within certain patients and are consistent with a progression of subclass responses at sites of infection and dis-

ease. An intriguing facet of host responses in periodontitis has been the relationship between local

and systemic antibodies. The general paradigm exists that gingival crevicular fluid is comprised primarily of a serum transudate at the site of inflammation (37,42). However, other findings are consistent with systemic antibody being a manifestation of the local antibody responses in the gingival tissues and that a portion of serum antibody may be derived from local gingival responses to this infection. The

studies described above document some characteristics of the local immune response and their relationship to infection and clinical presentation in

early-onset periodontitis patients, particularly within A. actinomycetemcomitans infected periodontitis patients. The findings are consistent with the potential to utilize antibodies or local mediators in gingival crevicular fluid as adjuncts in the diagnosis of periodontitis and in defining the mechanisms of disease.

A. actinomycetemcomitans is not always found in localized early-onset periodontitis patients (108,119, 150,184,227,327). Furthermore, A. actinomycetemcomitans has been found in patients with no clinical attachment loss (47,75,209). Leukotoxin production, a factor considered crucial in the virulence attributes of A. actinomycetemcomitans, varies among strains and has been found in A. actinomycetemcomitans isolated from localized early-onset periodontitis rather than adult periodontitis or healthy patients (361). A study in Denmark, however, found no occurrence of the high leukotoxin-producing strains of A. actinomycetemcomitans in their

population of localized early-onset periodontitis patients (123). In summary A. actinomycetemcomitans does not seem to be found in all cases of localized

early-onset periodontitis. Nevertheless, clinical studies carried out observing effect of treatment with A. actinomycetemcomitans load have linked failure of treatment with inability to reduce the subgingival load of A. actinomycetemcomitans (114,214).

Inflammatory mediators

It has been demonstrated that sera from localized

early-onset periodontitis patients with decreased

24


neutrophil function can modify the activity of healthy neutrophils (2). The serum factors responsible for these effects have been identified as being inflammatory mediators such as prostaglandins and cytokines, which have correlated with periodontal disease (208). Adherent mononuclear cells from lo-

calized early-onset periodontitis patients have been found to secrete higher levels of prostaglandin E2

and tumor necrosis factor-a than generalized early- onset periodontitis patients or healthy controls (274). More recently, insulin-dependent diabetic patients with periodontitis have been reported to

manifest excessive levels of inflammatory mediators (256-258). It has been suggested that these observations indicate an underlying immune defect of monocytes in certain diagnostic groups (222).

Early investigations of genetic polymorphisms of cytokines have attempted to identify a possible link between different periodontal categories and specific alleles related to an ability to overproduce cytokines (57,159). In addition, a fartily linkage study of early-onset periodontitis suggested that the COX-1

enzyme system, which acts as a catalyst for the breakdown of arachidonic acid, one of the products of which is prostaglandin E2, may be associated with susceptibility to early-onset periodontitis (89).

Neutrophil chemotactic defects

A further host defense defect that may increase susceptibility to early-onset periodontitis is neutrophil chemotactic defects. A high percentage of members of families with a background of localized early-onset periodontitis have been reported to show abnormal neutrophil chemotaxis (323). Whether the trait is due to an intrinsic defect of neutrophils or is

caused by extrinsic factors in the sera and thus is

environmentally influenced remains controversial (2). The neutrophil chemotactic defect is not noted in all populations exhibiting localized early-onset periodontitis, and many families with localized

early-onset periodontitis do not show evidence of the defect (152,153). Decreased neutrophil chemotaxis in localized early-onset periodontitis has been linked to a reduced receptors density for chemoattractants such as N-formyl-methyl-leucyl-phenyl-

alanine, complement fragment C5a, leukotriene B,

and interleukin-8 (49,52). De Nardin (53) found that decreased N-formyl-methyl-leucyl-phenylalanine binding which was related to variations in N-formyl-

methyl-leucyl-phenylalanine receptor DNA in local-

ized early-onset periodontitis patients and healthy

controls. These preliminary results indicate a poss-

ible role for a N-formyl-methyl-leucyl-phenylalanine

receptor alteration in neutrophil chemotaxis. How- ever, many molecules related to neutrophil dysfunction in localized early-onset periodontitis patients belong to the same family of surface receptors and have similar morphogenic characteristics. De Nardin (53) hypothesized that a common underlying defect, at either the cell surface receptor or post-receptor level may contribute to the alteration in neutrophil function seen in patients with localized early-onset periodontitis.

Humoral immune aspects

Lu et al. (185) have shown that in certain populations serum IgG2 levels are increased in localized early-onset periodontitis patients, but not generalized early-onset periodontitis, adult periodontitis or periodontally healthy individuals. An interesting association is that black subjects in general have higher levels of IgG2 than whites and the incidence of localized early-onset periodontitis among black individuals has been shown to be up to 15 times greater than among whites (182). It thus appears that localized early-onset periodontitis patients, particularly black individuals, have a tendency to produce increased levels of IgG2 (310).

A family study of early-onset periodontitis indicated the possibility of genetic transmission of IgG2 levels (193). Smoking has been found to suppress the IgG2 response in adult periodontitis and generalized early-onset periodontitis patients and is associated with more severe destruction but this does not appear to occur in localized early-onset periodontitis patients (310). Furthermore, the paradoxically lower IgG2 titers seen in black generalized early-onset periodontitis smokers may be specific to A. actinomycetemcomitans and the IgG2 response against antigens of other bacteria may not be lower in generalized early-onset periodontitis subjects who smoke (306).

Fc receptors

It has been suggested that the subclass IgG2 is a poor opsonin because it fixes complement less effectively than IgG3 and IgGI and binds weakly to Fc receptors on the surface of phagocytes, which tends to complicate theories of protective effects afforded by high

anti-A. actinomycetemcomtians titers of IgG2 in black localized early-onset periodontitis patients (108,310). However, a further aspect of the immune

response may yet explain this phenomenon. Fc re-

25

Kinane et al.

ceptors are the crucial link between the humoral and inflammatory components of the host defense system. It is possible that Fcy receptors III may prepare polymorphonuclear leukocytes for phagocytosis mediated by Fcy receptors II (343).

It has been reported that the numbers of Fcy receptors II and Fcy receptors III on peripheral blood

neutrophils of localized early-onset periodontitis patients are within the normal range (172). There may be local suppression of Fcy receptor expression and/ or the increased production of Fcy-binding proteins by periodontal pathogens (130).

Conclusion

The various aspects of the host response that may contribute to peridontal disease have been covered in this chapter as have several aspects that may be

specifically relevant to early-onset periodontitis, in-

cluding aspects of the innate, inflammatory and im-

mune defense systems. In summary, the differences in the causation and pathogenesis of adult and early- onset forms of periodontitis are not yet sufficiently elucidated (322). However, multiple specific host defense features are emerging in the early-onset forms

of periodontitis that may have a genetic basis and that may serve to differentiate the different forms of periodontitis and may have utility in terms of screening, diagnosis and therapy.

References

1. Abe T, tiara Y, Aono M. Penetration, clearance and retention of antigen en route from the gingival sulcus to the draining lymph node of rats. J Periodontal Res 1991: 26: 429-439.

2. Agarwal S, Huang JP, Piesco NP, Suzuki JB, Riccelli AE,

Johns LP Altered neutrophil function in localized juvenile

periodontitis: intrinsic or induced? J Periodontol 1996: 67:

337-344.

3. Aiba T, Akeno N, Kawane T, Okamoto H, Horiuchi N. Ma- trix metalloproteinases-1 and -8 and TIMP-1 mRNA levels in normal and diseased human gingivae. Eur J Oral Sci 1996: 104: 562-569.

4. Albandar JM, Brown LI, Genco RI, Me H. Clinical classification of periodontitis in adolescents and young adults. I Periodontol 1997: 68: 545-555.

5. Aldred MI, Bartold PM. Genetic disorders of the gingivae

and periodontium. Periodontol 2000 1998: 18: 7-20.

6. Alexander MB, Damoulis PD. The role of cytokines in the pathogenesis of periodontal disease. Curr Opin Peri- odontol 1994: 39-53.

7. Altman LC, Page RC, Ebersole IL, Vandesteen EG. Assess-

ment of host defences and serum antibodies to suspected

periodontal pathogens in patients with various types of periodontitis. J Periodontal Res 1982: 17: 495-497.

8. Amino N, Miyai K, Yamamoto T. Transient recurrence of hyperthyroidism after delivery in Graves' disease. J Clin Endocrinol Metabol 1977: 44: 130-136.

9. Anavi Y, Lerman P, Mintz S, Kiviti S. Idiopathic familial gingival fibromatosis associated with mental retardation, epilepsy and hypertrichosis. Dev Med Child Neurol 1989: 31: 538-542.

10. Aoyagi T, Sugawara-Aoyagi M, Yamazaki K, Hara K. In- terleukin 4 (IL-4) and IL-6-producing memory T cells in

peripheral blood and gingival tissues in periodontitis patients with high serum antibody titres to Porphyromonas

gingivalis. Oral Microbiol Immunol 1995: 10: 304-310.

11. Asikainen S, Lai C-H, Alaluusua S, Slots J. Distribution of Actinobacillus actinomycetemcomitans serotypes in periodontal health and disease. Oral Microbiol Immunol 1991: 6: 115-118.

12. Attström R, Egelberg J. Emigration of blood neutrophils and monocytes into the gingival crevice. J Periodontal Res 1970: 5: 48-55.

13. Attström R, Van der Velden U. Summary of session 1. In: Lang N, Kar-ring T, ed. Proceedings of the Ist European Workshop in Periodontology. Berlin: Quintessence, 1993: 120-126.

14. Aukhil 1, Lopatin DE, Syed SA, Morrison EC, Kowalski CJ. The effects of periodontal therapy on serum antibody (IgG) levels to plaque microorganisms. I Clin Periodontol 1988: 15: 544-550.

15. Baab DA, Page RC, Ebersole JL, Williams BL, Scott CR. Laboratory studies of a family manifesting premature exfoliation of deciduous teeth. I Clin Periodontol 1986: 13: 677-683.

16. Bashirelahi N, Organ RI, Bregquist JJ. Steroid binding protein in human gingiva. I Dent Res 1977: 56: A125.

17. Bendtzen K. Cytokines and natural regulators of cytokines. Immunol Lett 1994: 43: 111-123.

18. Benjamini E, Sunshine G, Leskowitz S. Immunology. A

short course. 3rd edn. New York: Wiley-Liss, 1996. 19. Beuscher U, Rausch P, Otterness G, Tollinghoff M. Tran-

sition from interleukin 1 beta to IL-1 alpha production during maturation of inflammatory macrophages in vivo. J Exp Med 1992: 175: 1793-1797.

20. Bevilacqua M, Pober J, Wheeler M, Cotran R, Gimbrone M Jr. Interleukin-1 acts on cultured human vascular endothelium to increase the adhesion of polymorphonuclear leukocytes, monocytes, and related leukocyte cell lines. J Clin Invest 1985: 76: 2003-2011.

21. Bimstein E, Ebersole JL. Serum antibody levels to oral microorganisms in children and young adults with relation to the severity of gingival disease. Pediatr Dent 1991: 13: 267-272.

22. Birkedal-Hansen H. Roles of cytokines and inflammatory

mediators in tissue destruction. J Periodontal Res 1993:

28: 500-510.

23. Bolstad Al, Kristoffersen T, Olsen I, Preus HR, Jensen HB, Vasstrand EN, Bakken V. Outer membrane proteins of Actinobacillus actinomyceterncomitans and Haemophilus

aphrophilus studied by SDS-PAGE and immunoblotting. Oral Microbiol Immunol 1990: 5: 155-161.

24. Bragd L, Dahlen G, Wikström M, Slots J. The capability of Actinobacillus act inomycetemcomitans, Bacteroides gingivalis and Bacteroides intermedius to indicate progressive

26


periodontitis: a retrospective study. J Clin Periodontol 1987: 14: 95-99.

25. Bragd L, Dahlen G, Wikström M, Slots J. The capability of Actinobacillus actinomycetemcomitans, Bacteroides gingivalis and Bacteroides intermedius to indicate progressive periodontitis; a retrospective study. J Clin Periodontol 1987: 14: 95-99.

26. Brecx M, Frölicher 1, Gehr P, Lang NP. Stereological observations on long term experimental gingivitis in man. J Clin Periodontol 1988: 15: 621-627.

27. Brecx M, Gautchi M, Gehr P, Lang N. Variability of histo-

logic criteria in clinically healthy human gingiva. J Peri-

odontal Res 1987: 22: 468-472.

28. Califano JV, Schenkein HA, Tew JG. Immunodominant

antigens of Actinobacillus actinomycetemcomitans serotype b in early-onset periodontitis patients. Oral Micro- biol Immunol 1992: 7: 65-70.

29. Campochiaro A. Cytokine production by retinal pigmented epithelial cells. Int Rev Cytol 1993: 146: 75-82.

30. Cappelli D, Ebersole IL, Komman KS. Early-onset periodontitis in hispanic adolescents associated with A. acti-

nomycetemcomitans. Community Dent Oral Epidemiol

1994: 22: 116-121.

31. Carswell E, Old 1, Kassel L, Green S, Fiore N, Williamson

B. An endotoxin-induced serurrr factor that causes ne-

crosis of tumors. Proc Natl Acad Sci USA 1975: 72: 3666-

3670. 32. Celenligil H, Ebersole JL. Analysis of serum antibody re-

sponses to periodontopathogens in early-onset peri-

odontitis from different geographical locations. J Clin

Periodontol 1998: 25: 994-1002.

33. Celenligil fl, Ebersole JL. Periodontal findings and systemic antibody responses to oral microorganisms in

Behcet's disease. J Clin Periodontol 1999: 70: 1449-1456.

34. celenligil El, Ebersole JL. Periodontal status and serum

antibody responses to oral microorganisms in Sjögren's

syndrome. I Periodontol 1998: 69: 571-577.

35. C elenligil H, Kansu E, Eratalay K. Juvenile and rapidly progressive periodontitis. Peripheral blood lymphocyte

populations. J Clin Periodontol 1990: 17: 207-210. 36.4elenligil 11, Kansu E, Ruacan S, Eratalay K, Caglayan G.

In situ characterization of gingival mononuclear cells in

rapidly progressive periodontitis. J Periodontol 1993: 64:

120-127.

37. Challacombe SJ, Russell MW, Hawkes J. Passage of intact lgG from plasma to the oral cavity via crevicular fluid. Clin E: xp Immunol 1978: 34: 417-422.

38. Chen CC, Chang KL, Huang JF Huang IS, Tsai CC. Corre-

lation of interleukin-1 beta, interleukin-6 and peri-

odontitis. Kao Hsiung J Med Sci 1997: 13: 609-617.

39. Chen EIA, Johnson BD, Sims TI, Darveau RP, Moncla B(, Whitney CW, Engel D, Page RC. Humoral immune responses to Porphyromonas gingivalis before and following therapy in rapidly progressive periodontitis patients. I Periodontol 1991: 62: 781-791.

40. Chung C-P, Lee Y-K, Choi S-M, Nisengard RI. Antibodies

to Actinobacillus actinomycetemcomitans in a Korean

population. J Periodontol 1986: 57: 510-515.

41. Chung PC, Nisengard RJ, Slots J, Genco RI. Bacterial JgG

and IgM antibody titers in acute necrotizing ulcerative

gingivitis. J Periodontol 1983: 54: 557-562.

42. Cimasoni G. Crevicular fluid updated. Monogr Oral Sci 1983: 12: iii-vii, 1-152.

43. Clemens M, Read AP, Brown T, ed. Cytokines. Oxford: Bios Scientific Publishers, 1991.

44. Cobb CM, Singla 0, Feil PH, Theisen FC, Schultz P Com- parison of NK-cell (Leu-7' and Leu-1lb') populations in clinically healthy gingivae, chronic gingivitis and chronic adult periodontitis. I Periodontal Res 1989: 24: 1-7.

45. Cockey GH, Boughman JA, Harris EL, Hassell TM. Genetic control of variation in human gingival fibroblast proliferation rate. In Vitro Cell Dev Biol 1989: 25: 255-258.

46. Cohen DW, Shapiro J, Friedman L, Kyle GC, Franklin S. A longitudinal investigation of the periodontal changes during pregnancy and fifteen months post-partum. I Peri-

odontol 1971: 42: 653-657. 47. Dahlen G, Manji F, Baelum V, Fejerskov O. Black-pig-

mented Bacteroides species and Actinobacillus actinomycetemcomitans in subgingival plaque of adult Kenyans. J Clin Periodontol 1989: 16: 305-310.

48. Damle NK, Doyle W. Ability of human T lymphocytes to adhere to vascular endothelial cells and to augment endothelial permeability to macromolecules is linked to their state of posthymic maturation. J Immunol 1990: 144: 1233-1240.

49. Daniel MA, Van Dyke TE. Alterations in phagocyte function and periodontal infection. J Periodontol 1986: 67: 1070-1075.

50. Danielsen B, Wilton JMA, Baelum V, Johnson NW, Fejer-

skov O. Serum immunoglobulin G antibodies to Porphyr-

omonas gingivalis, Prevotella intermedia, Fusobacterium

nucleatum and Streptococcus sanguis during experimental gingivitis in young adults. Oral Microbiol Immunol 1993: 8: 154-160.

51. Dayer J, Beutler B, Cerami A. Cachectin/tumor necrosis factor stimulates collagenase and prostaglandin E, production by human synovial cells and dermal fibroblasts. J Exp Med 1985: 162: 2163-2168.

52. De Nardin E, De Luca C, Levine MJ, Genco RJ. Antibodies directed to the chemotactic factor receptor detect differ-

ences between chemotactically normal and defective

neutrophils from LJP patients. I Periodontol 1990: 61: 609-617.

53. De Nardin E. The molecular basis for neutrophil dysfunc-

tion in early-onset periodontitis. J Periodontol 1996: 67: 345-354.

54. De Waal Malefyt R, Yssel H, de Vries J. Direct effects of IL- 10 on subsets of human CD4 `T cell clones and resting T cells. Specific inhibition of IL-2 production and proliferation. J Immunol 1993: 150: 4754-4765.

55. De Waal Malefyt R, Yssel H, Roncarolo MG, Spits H, de Vries 1. Interleukin-10. Curr Opin Immunol 1992: 4: 314-320.

56. Del Prete G, De Carli M, Almerigogna F, Giudizi M, Biagi-

otti R, Romagnani S. Human IL-10 is produced by both helper (Thl) and Type 2 helper (Th2) T cell clones and inhibits their antigen-specific proliferation and cytokine

production. J Immunol 1993: 150: 353-360. 57. Diehl SR, Wang YE Beck JD, Gao Y, Khanna A, Gu X, Bur-

meister JA, Schenkein HA. lnterleukin-1 genotypes and

risk of early onset periodontitis -a family based study of linkage disequilibrium. I Dent Res 1998: 77: 717 (ab-

stract). 58. Dzink J, Socransky S, Ebersole J, Frey D. ELISA and con-

ventional techniques for identification of black-pigmented Bacteroides isolated from periodontal pockets. Periodontal Res 1983: 18: 369-374.

27

Kinane et al.

59. Ebersole IL, Cappelli D, Sandoval M-N, Steffen MJ. Anti-

gen specificity of serum antibody in A. actinomycetemcomitans-infected periodontitis patients. I Dent Res 1995: 74: 658--666.

60. Ebersole JL, Cappelli D, Sandoval M-N. Subgingival distribution of A. actinomycetemcomitans in periodontitis. J Clin Periodontol 1994: 21: 65-75.

61. Ebersole II., Cappelli D, Steffen MJ. Characteristics and utilization of antibody measurements in clinical studies of periodontal disease. J Periodontol 1992: 63: 1110-1116.

62. Ebersole JL, Holt SC, Cappelli D, Kesavalu L. Significance

of systemic antibody responses in the diagnostic and mechanistic aspects of progressing periodontitis. In: Hamada S, Holt SC, McGhee JR, ed. Periodontal disease:

pathogens & host immune responses. Chicago: Quintess-

ence Publishing Co., 1991: 343-357. 63. Ebersole IL, Holt SC. Immunological procedures for diag-

nosis and risk assessment in periodontal diseases. In:

Johnson NW, ed. Periodontal diseases: markers of disease

susceptibility and activity. Cambridge: Cambridge Univer-

sity Press, 1991: 203-227. 64. Ebersole JL, Sandoval M-N, Steffen MI, Cappelli D. Serum

antibody in A. actinomycetemcomitans-infected patients

with periodontal disease. Infect Immun 1991: 59: 1795-

1802. 65. Ebersole JL, Singer RE, Steffensen B, Filloon T, Kornman

KS. Inflammatory mediators and immunoglobulins in

GCF from healthy, gingivitis, and periodontitis sites. I

Periodontal Res 1993: 28: 543-546.

66. Ebersole JL, Taubman MA, Smith DI, Frey DE, Haffajee

AD, Socransky SS. Human serum antibody responses to

oral microorganisms IV. Correlation with homologous in-

fection. Oral Microbiol Immunol 1987: 2: 53-59.

67. Ebersole JL, Taubman MA, Smith DJ, Genco RJ, Frey DE.

Human immune responses to oral microorganisms. I. As-

sociation of localized juvenile periodontitis (LIP) with

serum antibody responses to Actinobacillus actinomycetemcomitans. Clin Exp Immunol 1982: 47: 43-52.

68. Ebersole IL, Taubman MA, Smith DJ, Goodson IM. Gingi-

val crevicular fluid antibody to oral microorganisms I. Method of collection and analysis of antibody. J Peri-

odontal Res 1984: 19: 124-130. 69. Ebersole JL, Taubman MA, Smith DJ, Hammond BE Frey

DE. Human immune responses to oral microorganisms. 11. Serum responses to Actinobacillus actinomycetem-

comitans and the correlation with localized juvenile peri-

odontitis. J Clin Immunol 1983: 3: 321-331.

70. Ebersole IL, Taubman MA, Smith DI, Socransky SS. Hu-

moral immune responses and diagnosis of human periodontal disease. I Periodontal Res 1982: 17: 478-480.

71. Ebersole IL, Taubman MA, Smith DJ. Gingival crevicular fluid antibody to oral microorganisms. It. Distribution

and specificity of local antibody responses. I Periodontal

Res 1985: 20: 349-356.

72. Ebersole IL, Taubman MA, Smith DI. Local antibody re-

sponses in periodontal diseases. I Periodontol 1985: 56:

51-55.

73. Ebersole IL. Systemic humoral immune responses in peri-

odontal disease. Crit Rev Oral Biol Med 1990: 1: 283-331.

74. EEC-Clearinghouse on Oral Problems Related to HIV In-

fection and WHO Collaborating Center on Oral Manifes-

tations of the Human Immunodeficiency Virus. Classifi-

cation and diagnostic criteria for oral lesions in HIV infec-

tion. I Oral Pathol Med 1993: 22: 289-291.

75. Eisenmann AA, Eisenmann R, Sousa 0, Slots J. Microbio- logical study of localized juvenile periodontitis in Pan-

ama. J Periodontol 1983: 54: 712-713. 76. Engel D, Monzingo S, Rabinovitch P Clagett 1, Stone R.

Mitogen-induced hyperproliferation response of peripheral blood mononuclear cells from patients with severe generalised periodontitis: lack of correlation with proportions of T cells and T cell subsets. Clin Immunol Im-

m uno pathol 1984: 30: 374-386.

77. Enk A, Angeloni V, Udey M, Katz S. Inhibition of Langer- hans cell antigen-presenting function by IL-10. J Immun- ol 1993: 150: 4754-4765.

78. Enwonwu CO. Infectious oral necrosis (cancrum oris) in Nigerian children: a review. Community Dent Oral Epide- miol 1985: 13: 190-194.

79. Frank RM. Bacterial penetration in the apical pocket wall of advanced human periodontitis. I Periodontal Res 1980: 15: 563-573.

80. Fujihashi K, Kono Y, Beagley KW Yamamoto M, McGhee JR, Mestecky I, Kiyono H. Cytokines and periodontal dis-

ease: immunopathological role of interleukins for B cell responses in chronic inflamed gingival tissues. 1 Peri-

odontol 1993: 64(suppl 5): 400-406. 81. Fujihashi K, Yamamoto M, Hiroi T, Bamberg TV, McGhee

JR, Kiyono H. Selected Thl and Th2 cytokine mRNA expression by CD4(+) T cells isolated from inflamed human

gingival tissues. Clin Exp Immunol 1996: 103: 422-428. 82. Fujita S, Takahashi H, Okabe H, Ozaki Y, Hara Y, Kato I.

Distribtuion of natural killer cells in periodontal diseases:

an immunohistochemical study. I Periodontol 1992: 63: 686-689.

83. Fullmer H, Gibson W, Lazarus G, Bladen H, Whedon A. The origin of collagenase in periodontal tissues of Man. I Dent Res 1969: 48: 646-651.

84. Galbraith GMP, Steed RB, Sanders JJ, Pandey JP. Tumor necrosis factor alpha production by oral leukocytes: influence of tumor necrosis factor genotype. J Periodontol 1998: 69: 428-433.

85. Gangbar S, Overall M, McCulloch C, Sodek J. Identifi- cation of polymorphonuclear leukocyte collagenase and gelatinase activities in mouthrinse samples: correlation with periodontal disease activity in adult and juvenile periodontitis. J Periodontal Res 1990: 25: 257-267.

86. Garant PR, Mulvihill JE. The fine structure of gingivitis in the beagle. III. Plasma cell infiltration of the subepithelial connective tissue. J Periodontal Res 1972: 7: 161-172.

87. Garrison W, Nichols C. LPS elicited secretory response in

monocytes: altered release of PGE, but not IL-1 beta in patients with adult periodontitis. I Periodontal Res 1989: 24: 88-95.

88. Gemmell E, Feldner B, Seymour GI. CD45RA and CD45RO positive CD4 cells in human peripheral blood and periodontal disease tissue before and after stimulation with periodontopathic bacteria. Oral Microbiol Immunol 1992: 7: 84-88.

89. Gemmell E, Marshall RI, Seymour GI. Cytokines and prostaglandins in immune homeostasis and tissue de-

struction in periodontal disease. Periodontol 2000 1997: 14: 112-143.

90. Gemmell E, Seymour GJ. Cytokine profiles of cells extracted from humans with periodontal diseases. J Dent Res 1998: 77: 16-26.

91. Genco RJ, 7ambon JJ, Christersson LA. The origin of periodontal infections. Adv Dent Res 1988: 2: 245-259.

28


92. Genco RJ, Slots J. Host responses in periodontal disease. J Dent Res 1984: 63: 441-451.

93. Genco RJ, Van Dyke TE, Levine MJ, Nelson RD, Wilson ME. Kreshover lecture. Molecular factors influencing neutrophil defects in periodontal disease. I Dent Res 1986: 65: 1379-1391.

94. Genco RJ, Van Dyke TE, Park B, Ciminelli M, Horoszewicz

H. Neutrophil chemotaxis impairment in juvenile periodontitis: evaluation of specificity, adherence, deform-

ability, and serum factors. J Reticuloendothol Soc 1980:

28(suppl): Bls-91s. 95. Genco RJ, Zambon 11, Murray PA. Serum and gingival

fluid antibodies as adjuncts in the diagnosis of Actino- bacillus actinomyceterncomitans associated periodontal disease. I Periodontol 1985: 56(suppl): 41-50.

96. Genco Ill. Host responses in periodontal diseases: current concepts. J Periodontol 1992: 63: 338-355.

97. Gillett R, Cruchley A, Johnson NW. The nature of the inflammatory infiltrates in childhood gingivitis, juvenile

periodontitis and adult periodontitis: immunocytochemical studies using a monoclonal antibody to HLA Dr. J Clin Periodontol 1986: 13: 281-288.

98. Glickman I. Clinical periodontology. 3rd edn. Philadel-

phia: W. B. Saunders Co., 1966: 84-85. 99. Gmür R, Baehni PC. Serum immu' oglobulin G responses

to various Actinobacillus acitnomycetemcomitans serotypes in a young ethnographically heterogeneous peri-

odontitis patient group. Oral Microbiol Immunol 1997:

12: 1-10.

100. Gmür R, Hrodek K, Saxer OR Guggenheim B. Double-

blind analysis of the relation between adult periodontitis

and systemic host response to suspected periodontal

pathogens. Infect Immun 1986: 52: 768-776.

101. Golub L, Siegel K, Ramamurthy S, Mandel 1. Some characteristics of collagenase activity in gingival crevicular fluid

and its relationship to gingival diseases in humans. J Dent

Res 1976: 55: 1049-1057.

102. Gore EA, Sanders JJ, Pandey JP, Palesch Y, Galbraith GMP. Interleukin-1ß+3953 allele 2: association with disease status in adult periodontitis. I Clin Periodontol 1998: 25: 781-785.

103. Grbic JT, Singer RE, Jans 1{H, Celenti RS, Lamster IB. Im- munoglobulin isotypes in gingival crevicular fluid: possible protective role of IgA. J Periodontol 1995: 66: 55-61.

104. Grieve WG, Murray PA, Winkler JR. Humoral immune re-

sponse in 1JIV-associated periodontal disease. J Dent Res

1989: 68: 646.

105. Grieve WG. Humoral immune response in HIV-associated

periodontitis. Dentistry 1989: 89: 19-23.

106. Gunsolley JC, Burmeister JA, Tew JG, Best A, Ranney RR.

Relationship of serum antibody to attachment level patterns in young adults with juvenile periodontitis or gener-

alized severe periodontitis. J Periodontol 1987: 58: 314-

320.

107. Gunsolley IC, Quinn SM, Tew J, Gooss CM, Brooks CN,

Schenkein HA. The effect of smoking on individuals with minimal periodontal destruction. J Periodontol 1998: 69:

165-170.

108. Gunsolley IC, Ranney RR, Zambon JJ, Burmeister JA,

Schenkein IIA. Actinobacillus actinomycetemcomitans in families afflicted with periodontitis. I Periodontol 1990:

61: 643-648.

109. Gunsolley JC, Tew JG, Gooss CM, Burmeister JA, Schenk-

ein HA. Effects of race and periodontal status on antibody reactive with Actinobacillus actinomycetemcomitans strain Y4. J Periodontal Res 1988: 23: 303-307.

110. Haerian A, Adonogianaki E, Mooney J, Docherty J, Kinane D. Gingival crevicular stromelysin, collagenase and tissue inhibitor of metalloproteinases levels in healthy and dis-

eased sites. J Clin Periodontol 1995: 22: 505-509. 111. Haerian A, Adonogianaki E, Mooney 1, Manos A, Kinane

D. Effects of treatment on gingival crevicular collagenase, stromyelysin and tissue inhibitor of metalloproteinases and their ability to predict response to treatment. I Clin Periodontol 1996: 23: 83-91.

112. Haffajee A, Dzink J, Socransky S. Effect of modified Wid- man flap surgery and systemic tetracycline on the subgingival microbiota of periodontal lesions. J Clin Periodontol 1988: 15: 255-262.

113. Haffajee A, Socransky S, Dzink J, Taubman M, Ebersole J. Clinical, microbiological and immunological features of subjects with refractory periodontal diseases. I Clin Peri- odo ntol 1988: IS: 390-398.

114. Haffajee AD, Socransky SS, Ebersole IL. Clinical, microbiological and immunological features associated with the treatment of active periodontal lesions. J Clin Peri- odo ntol 1984: 11: 600-618.

115. Hall E, Ebersole JL, Prihoda TJ, Steffen MJ. Antigenic di-

versity in A. actinomycetemcomitans presumptive identification. J Dent Res 1993: 72: 244.

116. Hall ER, Falkler WA, Martin SA, Suzuki JB. The gingival immune response to Actinobacillus actinomycetemcomitans in juvenile periodontitis I Periodontol 1991: 62: 792- 798.

117. Hall ER, Martin SA, Suzuki JB, Falkler WA Jr. The gingival immune response to periodontal pathogens in juvenile periodontitis Oral Microbiol Immunol 1994: 9: 327-334.

118. Halliwell B, Gutteridge JMC. The importance of free radicals and catalytic metal ions in human disease. Mol As-

pects Med 1985: 8: 89-193. 119. Han N, Xiao X, Zhang L, Ri X, Zhang J, Tong Y, Yang M,

Xiao Z. Bacteriological study of juvenile periodontitis in China. I Periodontal Res 1991: 26: 409-414.

120. Hanazawa S, Amano S, Nakda K, Ohmori Y, Miyoshi T, Hirose K, Kitomo S. Biological characterization of in-

terleukin-1 like cytokine produced by cultured bone cells from newborn mouse calvaria. Calcif Tissue Int 1987: 141:

31-37. 121. Harri MP, Ojanotko AO. Progesterone metabolism in

healthy and inflamed female gingiva. J Steroidal Biochem 1978: 9: 826-835.

122. Hassell TM, Baehni P, Harris EL, Walker C, Gabbiani G, Geinoz A. Evidence for genetic control of changes in f-

actin polymerization caused by pathogenic microorganisms: in vitro assessment using gingival fibroblasts from human twins. I Periodontal Res 1997: 32: 90-98.

123. Haubek D, Poulsen K, Asikainen S, Kilian M. Evidence for

absence in northern Europe of especially virulent clonal types of Actinobacillus actinomycetemcomitan& I Clin Microbiol 1995: 3: 395-401.

124. Havarth L, Leonard EJ. Two neutrophil populations in hu-

man blood with different chemotactic activities: separation and chemoattractant binding. Infect Immun 1982: 36: 443-449.

125. Heath J, Gowen M, Meikle M, Reynolds J. Human gingival tissues in culture synthesis three metalloproteinases and

29

Kinane et al.

126

127.

128.

129.

130.

131.

132

a metalloproteinase inhibitor. I Periodontal Res 1982: 17: 183-190. Heath J, Saklatvala J, Meikle M, Atkinson S, Reynolds J. Periosteal fibroblasts synthesize a cytokine that stimulates bone resorption and demonstrates interleukin-1 activity after partial purification. Br Rheumatol 1985:

24(suppl): 136-139. Hedges S, Agace W, Svensson M, Sjögren C, Ceska M, Svanborg C. Uroepithelial cells are part of a mucosal cytokine network. Infect Immun 1994: 62: 2315-2321.

Hench PA. The potential reversibility of rheumatoid arthritis. Mayo Clin Proc 1949: 24: 167-178. Hendley TM, Steed RB, Galbraith GM. Interleukin- 1ß gene expression in human oral polymorphonuclear leukocytes. I Periodontol 1995: 66: 761-765. Hillestad M, Helgeland K, Tolo K. Fcy-binding bacteria in

periodontal lesions. Oral Microbiol Immunol 1996: 11:

242-247. Hillmans G, Hillmans B, Geurtsen W. Immunohistolog-

ical determination of interleukin-1 beta in inflamed hu-

man gingival epithelium. Arch Oral Biol 1995: 40: 353-

359. Hirano T. Interleukin-6. In: Thomson A, ed. The cytokine handbook. Ist edn. New York: Academic Press, 1991: 169-

190. 133. Flolm-Pedersen P, Löe H. Flow of gingival exudate as re-

lated to menstruation and pregnancy. I Periodontal Res

1967: 2: 13-30.

134. Holmstrup P, Westergaard J. Necrotizing periodontal dis-

ease. In: Lindhe 1, Karring T, Lang NP, eds. Clinical period-

ontology and implant dentistry. 3rd edn. Copenhagen:

Munksgaard, 1997: 258-278.

135. Horning GM, Hatch CL, Lutskus J. The prevalence of peri-

odontitis in a military treatment population. J Am Dent

Assoc 1990: 121: 616-622.

136. Horton JE, Raisz LG, Simmons HA, Oppenheim JJ,

Mergenhagen SE. Bone resorbing activity in supernatant fluid from cultured human peripheral blood leukocytes.

137.

138.

139.

140.

141.

Science 1972: 177: 793-795.

Howells Gl_ Cytokine networks in destructive periodontal disease. Oral Dis 1995: 1: 266-270. Hugoson A. Gingival inflammation and female sex hor-

mones. A clinical investigation of pregnant women and experimental studies in dogs. J Periodontal Res 1970: 5(suppl): 1-18. Ingman T, Sorsa T, Konttinen Y, Saarai I-I, Lindy 0, Suoma- lainen K. Salivary collagenase, elastase- and trypsin-like proteases as biochemical markers of periodontal tissue destruction in adult and localized juvenile periodontitis. Oral Microbiol lmmunol 1993: 8: 298-305. Ingman T, Tervahartiala T, Ding Y, Tschesche 11, Haerian

A, Kinane D. Matrix metalloproteinases and their inhibi-

tors in gingival crevicular fluid and saliva of periodontitis

patients. J Clin Periodontol 1996: 23: 1127-1132. Ishihara Y, Nishihara T, Maki E, Noguchi T, Koga T. Role

of interleukin-1 and prostaglandin in in vitro bone resorption induced by Actinobacillus actinomycetemcomit-

ans lipopolysaccharide. ) Periodontal Res 1991: 26: 155-

160. 142. Jacob C, McDevitt U. Tumour necrosis factor-a in murine

autoimmune "lupus" nerhritis. Nature 1998: 331: 356-358.

143. Jacobs E, Willer TF, Nauman RK. Detection of elevated

serum antibodies to T}eponema denticola in humans with

144.

145

146.

147

advanced periodontitis by an enzyme-linked immuno-

sorbent assay. J Periodontal Res 1982: 17: 145-153. Iandinski JI, Stashenko P, Feder LS, Leung CC, Peros WI, Rynar JE, Deasy MI. Localization of interleukin-1 ß in hu-

man periodontal tissue. I Periodont 1991: 62: 36-43. Jinquan T, Gronhoj Larsen C, Gesser B, Matsushima K, Thestrup-Pedersen T. Human IL-10 is a chemoattractant for CD8` T lymphocytes and an inhibitor of IL-8 induced CD4' T lymphocyte migration. J Immunol 1993: 151: 4545-4551. Johnson V, Johnson BD, Sims Ti, Whitney CW, Moncla BJ, Engel D, Page RC. Effects of treatment of antibody titer to Porphyromonas gingivalis in gingival crevicular fluid of patients with rapidly progressive periodontitis. 1 Peri-

odo ntol 1993: 64: 559-565.

Kabashima H, Nagata K, Hashiguchi 1, Toriya Y, lijima T, Maki K, Maeda K. Interleukin-1 receptor antagonist and interleukin-4 in gingival crevicular fluid of patients with inflammatory periodontal disease. I Oral Pathol Med 1996: 25: 449-455.

148.

149.

150.

151.

152.

153.

154.

155.

156.

157.

158.

159

Katz J, Goultschin J, Benoliel R, Schlesinger M. Peripheral T lymphocyte subsets in rapidly progressive periodontitis. J Clin Periodontol 1988: 15: 266-268. Kaufmann AY. An oral contraceptive as an etiologic factor in producing hyperplastic gingivitis and a neoplasm of the pregnancy tumor type. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1969: 28: 666-670. Kim KI, Son S, Chung CP, Kim DK. Longitudinal monitoring for disease progression of localised juvenile periodontitis. J Periodontol 1992: 63: 806-811. Kinane DF, Lindhe J. Pathogenesis of periodontitis. In: Lindhe 1, Karring T, Lang NP, ed. Clinical periodontology and implant dentistry. 3rd edn. Copenhagen: Munks- gaard, 1997: 89-225. Kinane DF, Cullen CF, Johnston FA, Evans CW. Neutrophil chemotactic behaviour in patients with early-onset forms of periodontitis. I. Leading front analysis in Boyden chambers. J Clin Periodontol 1989: 16: 242-246. Kinane DF, Cullen CF, Johnston FA, Evans CW. Neutrophil

chemotactic behaviour in patients with early-onset forms

of periodontitis. II. J Clin Periodontal 1989: 16: 242-246. Kinane DF, Johnstone FA, Evans CW. Depressed helper-to-

surpressor T cell ratios in early onset forms of periodontal disease. J Periodontal Res 1989: 24: 161-164. Kinane DF, Lappin DF, Koulouri 0, Buckley A. Humoral immune responses in periodontal disease may have mucosal and systemic immune features. Clin Exp Immunol

1999: 115: 534-541.

Kjeldsen M, Holmstrup P. Bendtzen K. Marginal periodontitis and cytokines: a review of the literature. 1 Peri-

odontol 1993: 64: 1013-1022. Koh H, Whitney C, Johnson B, Houston L, Moncla B, Engel

D, Page R. Serum IgG subclass response to A. actinomycetemcomitans in rapidly progressive periodontitis. I Dent

Res 1991: 70: 284 (abstr).

Kopp W. Density and localisation of lymphocytes with natural killer (NK) cell activity in periodontal biopsy

specimens from patients with severe periodontitis. I Clin Periodontol 1988: 15: 595-600. Kornman KS, Crane A, Wang HY, di Giovine FS, Newman

MG, Pirk FW, Wilson TG Jr, Higginbottom FL, Duff GW.

The interleukin-1 genotype as a severity factor in adult periodontal disease. 1 Clin Periodontol 1997: 24: 72-77.

30


160.

161

Kornman KS, Loesche WI. The subgingival microflora during pregnancy. J Periodontal Res 1980: 15: 111-122. Kryshtalskyj E, Sodek J, Ferrier J. Correlation of colla-

genolytic enzymes and inhibitors in gingival crevicular fluid with clinical and microscopic changes in experimental periodontitis in the dog. Arch Oral Biol 1986: 31: 21-

31.

177. Lindhe 1, Birch I, Br'nemark PI. Vascular proliferation in pseudopregnant rabbits. J Periodontal Res 1968: 3: 12-20.

178. Lindhe J, Hamp SE, We H. Plaque induced periodontal disease in beagle dogs. A four year clinical roentgenographical and histometric study. I Periodontal Res 1975: 10: 243-255.

179. Ustgarten MA, Lai CH, Evian Cl. Comparative antibody titres to Actinobacillus actinomycetemcomitans in juvenile periodontitis, chronic periodontitis and periodontally healthy subjects. I Clin Periodontol 1981: 18: 155-164.

180. Listgarten MA. Electron microscopic observations on the bacterial flora of acute necrotising ulcerative gingivitis. J Periodontol 1965: 36: 328-339.

181. Listgarten MA. Structure of the microbial flora associated with periodontal health and disease in man. A light and electron microscopic study. J Periodontol 1976: 47: 1-8.

182. We H, Brown LJ. Early onset periodontitis in the United States of America. 1 Periodontol 1991: 62: 608-616.

183. tile H, Theilade E, Jensen S. Experimental gingivitis in man. J Periodontol 1965: 36: 177-187.

184. Lopez NJ, Mellado IC, Giglio MS, Leighton GX- Occur-

rence of certain bacterial species and morphotypes in juvenile periodontitis in Chile. J Periodontol 1995: 66: 559-567.

185. Lu H, Wang M, Gunsolley JC, Schenkein HA, Tew 1G. Serum immunoglobulin G subclass concentrations in

periodontally healthy and diseased individuals. Infect Im-

mun 1994: 62: 1677-1682. 186. Luger T, Stadler B, Katz S, Oppenheim J. Epidermal cell

(keratinocyte) derived thymocyte activating factor. I Im-

munol 1981: 127: 1493-1498. 187. Lynn BD. "The Pill" as an etiologic agent in hypertropic

gingivitis. Oral Surg Oral Med Oral Pathol Oral Radiol En- dod 1969: 24: 333-334.

188. Mackler BE Frostad KB, Robertson RB, Levy BM. Im-

munoglobulin-bearing lymphocytes and plasma cells in human periodontal disease. I Periodontal Res 1977: 12: 37-45.

189. Mackler BE Waldrop TC, Schur P, Robertson RB, Levy BM. IgG subclasses in human periodontal disease I. Distri- bution and incidence of IgG subclasses bearing lympho-

cytes and plasma cells. I Periodontal Res 1978: 13: 109-119. 190. Mahanonda R, Seymour G, Powell L, Good M, Halliday J.

Effect of initial treatment of chronic inflammatory periodontal disease on the frequency of peripheral blood T- lymphocytes specific to periodontopathic bacteria. Oral Microbiol Immunol 1991: 6: 221-227.

191. Mandell RL, Ebersole LI, Socransky SS. Clinical immunol-

ogic and microbiologic features of active disease sites in juvenile periodontitis. J Clin Periodontol 1987: 14: 534- 540.

192. Manhart SS, Reinhardt RA, Payne JB, Seymour GL, Gem- mel E, Dyer IK, Petro TM. Gingival cell IL-2 and IL-4 in early onset periodontitis. I Periodontol 1994: 144: 662- 666.

193. Marazita ML, Lu H, Cooper ME, Quinn SM, Zhang 1, Bur- meister JA, Califano JV, Pandey JP, Schenkein HA, Tew IG. Genetic segregation analyses of serum IgG2 levels. Am I Hum Genet 1996: 58: 1042-1049.

194. Mariotti A. Sex steroid hormones and cell dynamics in the periodontium. Crit Rev Oral Biol Med 1994: 5: 27-53.

195. Marshall-Day CD. The epidemiology of periodontal disease. I Periodontol 1951: 22: 13-22.

162.

163.

164.

165.

166.

167.

168.

169.

170.

171.

172.

173.

174

Kryshtalskyj E, Sodek J. Nature of collagenolytic enzyme and inhibitor activities in crevicular fluid from healthy

and inflamed periodontal tissue of beagle dogs. I Peri- odontal Res 1987: 22: 264-269. Kubota T, Matsuki Y, Nomura T, Hara K. In situ hybridization study on tissue inhibitors of metalloproteinases (TIMPs) mRNA-expressing cells in human inflamed gingival tissue. I Periodontal Res 1997: 32: 467-472. Kubota T, Nomura T, Takahashi T, Hara K. Expression of

mRNA for matrix metalloproteinases and tissue inhibitors

of metalloproteinases in periodontitis-affected human

gingival tissue. Arch Oral Biol 1996: 41: 253-262.

Lai CH, L. istgarten M, Shirakawa M, Slots J. Bacteroides forsythus in adult gingivitis and periodontitis. Oral Micro- biol Immunol 1987: 2: 152-157. Lai C-H, L. istgarten MA, Evian Cl, Dougherty P. Serum IgA

and IgG antibodies to Treponema vincentii and 7Yepone-

ma denticola in adult periodontitis, juvenile periodontitis

and periodontally healthy subjects. J Clin Periodontol

1986: 13: 752-757. Lally ET, Baehni PC, McArthur WP. Local immunoglobulin

synthesis in periodontal disease. I Periodontal Res 1980:

15: 159-164. Lappin DP Koulouri 0, Radvar M, Hodge P, Kinane DE

Relative proportions of mononuclear cell types in peri-

odontal lesions analysed by immunohistochemistry.

Clin Periodontol 1999: 26: 183-189.

Larivee 1, Sodek 1, Ferrier I. Collagenase and collagenase inhibitor activities in crevicular fluid of patients receiving treatment for localized juvenile periodontitis. I Peri-

odontal Res 1986: 21: 702-715.

Le 1, Vilcek J. Tumor necrosis factor and interleukin-1:

cytokines with multiple overlapping biological activities. L. ab Invest 1987: 56: 234-248.

Lee W, Aitken S, Sodek 1, McCulloch C. Evidence of a di-

rect relationship between neutrophil collagenase activity

and periodontal tissue destruction in vivo: role of active

enzymes in human periodontitis. J Periodontal Res 1995:

30: 23-33.

L. eino L, 1-lurttia HM, Sorvajärvi K, Sewon LA. Increased

respiratory burst activity is associated with normal expression of IgG-Fc-receptors and complement receptors in peripheral neutrophils from patients with juvenile periodontitis. J Periodontal Res 1994: 29: 179-184. [. iljenberg B, [. indhe 1, Berglundh T, Dahlen G. Some

microbiological, histopathological and immunohisto-

chemical characteristics of progressive periodontal dis-

ease. I Clin Periodontol 1994: 21: 720-727. Lindemann RA. Bacterial activation of human natural killer cells: role of cell surface lipopolysaccharide. Infect

175.

176_

Immun 1988: 56: 1301-1308.

[. indhe J, Attström R, Björn AL Influence of sex hormones

on gingival exudation in dogs with chronic gingivitis. J Periodontol 1968: 3: 279-283. Lindhe 1, Attström R. Gingival exudation during the menstrual cycle. I Periodontal Res 1967: 2: 194-198.

31

Kinane et al.

196. Marthur A, Michalowicz BS. Cell-mediated immune system regulation in periodontal diseases. Crit Rev Oral Biol Med 1997: 8: 76-89.

197. Matsuki Y, Yamamoto T, Hara K. Detection of inflamma-

tory cytokine messenger RNA (mRNA) -expressing cells in human inflamed gingiva by combined in situ hybridiza-

tion and immunohistochemistry. Immunology 1992: 76: 42-47.

198. McCarthy D, Claffey N. Acute necrotising ulcerative gingivitis is associated with attachment loss. J Clin Periodontol

199

200

201.

202.

203.

204.

205.

206.

207.

208.

1991: 18: 776-779. McPhail LC, Harvarth L. Signal transduction in neutrophil oxidative metabolism and chemotaxis. In: Abramson IS, Wheeler JG, ed. The neutrophil. New York. Oxford Univer-

sity Press, 1993.

Meikle MC, Atkinson SJ, Ward RV, Murphy G, Reynolds JJ. Gingival fibroblasts degrade type I collagen films when stimulated with tumor necrosis factor and interleukin 1. Evidence that breakdown is mediated by metalloproteinases. J Periodontal Res 1989: 24: 207-213. Melvin WL, Sandifer IB, Gray JL. The prevalence and sex

ratio of juvenile periodontitis in a young racially mixed

population. I Periodontol 1991: 62: 330-334.

Meng FIX, Zheng LE T cells and T-cell subsets in peri-

odontal diseases I Periodontal Re% 1989: 24: 121-126.

Meyle J, Gonzalez 1. Systemic diseases predisposing to

periodontitis in children and adolescents. Periodontol

2000 2001: 26: 00-00. Miossec P, Cavender D, Ziff M. Production of interleukin

I by human endothelial cells. I Immunol 1986: 136: 248-

249. Miyazaki A, Kobayashi T, Suzuki T, Yoshie H, Hara K. Loss

of Fcy receptor and impaired phagocytosis of polymor-

phonuclear leukocytes in gingival crevicular fluid. I Peri-

odontal Res 1997: 32: 439-446.

Mochan E, Armor L, Sporer R. Interleukin I stimulation of plasminogen activator production in cultured gingival fibroblasts. J Periodontal Res 1988: 23: 28-32. Modlin RL, Nutman TB. Type 2 cytokines and negative immune regulation in human infections. Curr Opin Im-

munol 1993: 5: 511-517. Molvig J, Baek L, Christensen P, Manogue KR, Vlassara

If, Platz P, Nielsen LS, Svejgaard A, Nerup I. Endotoxin-

stimulated human monocyte secretion of interleukin 1,

tumour necrosis factor alpha, and prostaglandin E2 shows

stable interindividual differences. Scand I Immunol 1988:

27: 705-716. 209. Mombelli A, Gmur R, Frey J, Meyer J, Zee KY, Tam IOW,

Lo ECM, Di Renzo J, Lang NP, Corbet EE Actinobacillus

actinomycetemcomitans and Porphyromonas gingivalis in

young Chinese adults. Oral Microbiol Immunol 1998: 13:

231-237.

210. Mooney J, Kinane DF. Comparison of humoral immune

response to Porphyromonas gingivalis and Actinobacillus

actinomycetemcomitans in adult periodontitis and rapidly

progressive periodontitis. Oral Microbiol Immunol 1994:

9: 321-326.

211.. Moughal NA, Adonogianaki E, Thornhill Ml-, Kinane DF.

Endothelial cell leukocyte adhesion molecule-1 (ELAM-1)

and intercellular adhesion molecule-1 (ICAM-1) express-

ion in gingival tissue during health and experimentally induced gingivitis. I Periodontal Res 1992: 27: 623-630.

212. Mouton C, Hammond PG, Slots J, Genco RJ. Serum anti-

bodies to oral Bacteroides asaccharolyticus (Bacteroides

gingivalis): relationship to age and periodontal disease. Infect Immun 1981: 31: 182-192.

213. Murayama Y, Nagai A, Okamura K, Kurihara H, Nomura Y, Kokeguchi S, Kato K. Serum immunoglobulin G antibody to periodontal bacteria. Adv Dent Res 1988: 2: 339- 345.

214. Murray PA, Winkler JR, Peros WI, French CK, lippke JA. DNA probe detection of periodontal pathogens in HIV-

associated periodontal lesions. Oral Microbiol Immunol 1991: 6: 34-40.

215. Nakagawa S, Machida Y, Nakagawa T. Infection by Por- phyromonas gingivalis and Actinobacillus actinomycetemcomitans, and antibody responses at different ages in humans. J Periodontal Res 1994: 29: 9-16.

216. Nisengard RJ, Rogers RS. The treatment of desquamative gingival lesions. J Periodontol 1987: 58: 167-172.

217. Nomura T, Takahashi T, Hara K. Expression of TIMP-1, TIMP-2 and collagenase mRNA in periodontitis-affected human gingival tissue. J Periodontal Res 1993: 28: 354- 362.

218. Offenbacher S. Periodontal diseases: pathogenesis. Ann

Pe riodo ntol 1996: 1: 821-878. 219. Nutlay AG, Bhaskar SN, Weinmann JP, Budy AM. The ef-

fect of estrogen on the gingiva and alveolar bone of molars in rats and mice. J Dent Res 1954: 33: 115-127.

220. Nyman S. Studies on the influence of oestradiol and progesterone on granulation tissue. I Periodontal Res 1971:

suppl 7. 221. Offenbacher S, Heasman P, Collins I. Modulation of host

PGE2 secretion as a determinant of periodontal disease. J Pe riodo ntol 1993: 64: 432-444.

222. Offenbacher S. Periodontal diseases: pathogenesis. Ann Periodontol 1996: 1: 821-878.

223. Ogawa T, Tarkowski A, McGhee ML, Moldoveanu Z, Mes- tecky J, Hirsch HZ, Koopman WJ, Hamada S, McGhee JR, Kiyono H. Analysis of human IgG and IgA antibody secreting cells from localised chronic inflammatory tissue. J Immunol 1989: 142: 1140-1158.

224. Ohgi S, Kamoi K. A. actinomycetemcomitans in human

subgingival plaque and distribution of serotype. I Jpn As-

soc Periodontol 1990: 32: 450-469. 225. Ojanotko AO, Harri M-P Progesterone metabolism by rat

oral mucosa. IL The effect of pregnancy. J Periodontal Res 1982: 17: 196-201.

226. Okada H, Kida T, Yamagami H. Characterisation of the immunocompetent cells in human advanced periodontitis. J Periodontal Res 1982: 17: 472-473.

227. Okuda K, Naito Y, Ohta K, Fukomoto Y, Kimura Y, Ishika-

wa 1, Kinoshita S, Takazoe I. Bacteriological study of periodontal lesions in two sisters with juvenile periodontitis and their mother. Infect Immun 1984: 45: 118-121.

228. Okuda K, Saito A, Hirai K, Harano K, Kato T. Role of antibodies in periodontopathic bacterial infections. In: Genco R, ed. Molecular pathogenesis of periodontal disease. Washington, DC: Am Soc Microbiol, 1994: 257-265.

229. Onman GM, Allen RA, Bokach GM, Painter RG, Traynor AE, Sklar LA. Signal transduction and cytoskeletal activation in the neutrophil. Physiol Rev 1987: 67: 285-322.

230. Oppenheim J, Zachariae C, Mukaida N, Matsushima K. Properties of the novel proinflammatory supergene "in- tercrine" cytokine family. Annu Rev Immunol 1991: 9: 617-648.

32

of periodontitis in children and adolescents

231. Overall CM, Wiebkin 0, Thonard J. Demonstration of tissue collagenase activity in vivo and its relationship to inflammation severity in human gingiva. J Periodontal

Res 1987: 22: 81-88.

232. Page RC, Beck ID. Risk assessment for periodontal dis-

eases. Int Dent J 1997: 47: 61-87.

233. Page RC, Schroeder HE. Periodontitis in man and ani-

mals. A comparative review. Basel: Karger, 1982.

234. Page RC, Simpson DM, Ammons WE Host tissue response in chronic inflammatory periodontal disease IV. The periodontal and dental status of a group of aged great apes. I

Periodontol 1975: 46: 144-155.

235. Page RC, Sims T), Engel LD, Moncla BJ, Bainbridge B, Stray 1, Darveau RP. The immunodominant outer membrane antigen of A. actinomycetemcomitans is located in

the serotype-specific high-molecular-mass carbohydrate

moiety of lipopolysaccharide. Infect Immun 1991: 59:

3451-3462. 236. Page RC. Gingivitis. I Clin Periodontol 1986: 13: 345-355.

237. Patters MR, Niekrash CE, Lang NP. Assessment of comple-

ment cleavage in gingival fluid during experimental gingi-

vitis in man. I Clin Periodontol 1989: 16: 33-37.

238. Payne WA, Page RC, Ogilvie AL, Hall WB. Histopathologic features of the initial and early stages of experimental gingivitis in man. J Periodontal Res L975: 10: 51-64.

239. Pinchback IS, Gibbins JR, Hunter N. Vascular co-localization of proteolytic enzymes and proteinase inhibitors in

advanced periodontitis. J Pathol 1996: 179: 326-332. 240. Pitzalis C, Kingsley C, FIaskard D, Panayi G. The preferen-

tial accumulation of helper-induced T lymphocytes in in-

flammatory lesions: evidence for regulation by selective

endothelial and homotypic adhesion Eur I fmmunol 1988:

18: 1397-1404.

241. Powell 1. Mediators of inflammation and tissue destruc-

tive metabolism in gingival crevicular fluid (GCF). I Dent Res 1992: 72: 1848.

242. Preiss DS, Meyle J. Interleukin-15 concentration of gingi-

val crevicular fluid. I Periodontol 1994: 65: 423-428.

243. Purvis JA, Embery G, Oliver WM. Molecular size distri-

bution of proteoglycans in human gingival tissue. Arch

Oral Biol 1984: 29: 513-519.

244. Rams TE, Andriolo M, Feik D, Abel SN, McGivern TM,

Slots 1. Microbiological study of HIV-related periodontitis. J Periodontol 1991: 62: 74-81.

245. Reinhardt RA, McDonald TL, Bolton RW, DuBois LM, Kal-

dahl WB. IgG subclasses in gingival crevicular fluid from

active versus stable periodontal sites I Periodontol 1989:

60: 44--50. 246. Richards D, Rutherford RB. The effects of interleukin 1 on

collagenolytic activity and prostaglandin-E secretion by

human periodontal- ligament and gingival fibroblasts.

Arch Oral Biol 1988: 33: 237-243.

247. Richman M), Abarbanel AR. Effect of estradiol and di-

ethylstilbesterol upon the atropic human buccal mucosa

with a preliminary care report on the use of estrogens in the management of senile gingivitis. J Clin Endocrinol

Metab 1943: 3: 224-226.

248. Richman MI, Abarbanel AR. Effects of estradiol, testoster-

one, diethylstilbesterol and several of their derivatives

upon the human oral mucous membrane. J Am Dent As-

soc 1943: 30: 913--923.

249. Robertson P, Grupe 11, Taylor R, Shyu K, Fullmer if. The

effect of collagenase-inhibitor complexes on collagenolyt-

is activity of normal and inflamed gingival tissue. I Oral Pathol 1973: 2: 28-32.

250. Roitt IM. Essential immunology. 8th edn. Oxford: Blackwell Science, 1994.

251. Rosales C, Brown EJ. Neutrophil receptors and the modulation of the immune response. In: Abramson IS, Wheeler JG, ed. The neutrophil. New York: IRL PRess, 1993: 23-62.

252. Rossomando EF, Kennedy JE, Hadjmichael J. Tumour necrosis factor alpha in gingival crevicular fluid as a possible indicator of periodontal disease in humans. Arch Oral Biol 1990: 35: 431-434.

253. Rousset F, Garcia E, Defrance T, Peronne C, Vezzio N, Hsu D, Kastelein R, Moore K, Banchereau J. Interleukin-10 is

a potent growth and differentiation factor for activated human B lymphocytes. Proc Natl Acad Sci USA 1992: 89: 1890-1893.

254. Rudney JD, Michalowicz BS, Krig MA, Kane PK, Pihlstrom BL Genetic contributions to saliva protein concentrations in adult human twins. Arch Oral Biol 1994: 39: 513-517.

255. Saglie R, Newman MG, Carranza FA, Pattison GL. Bac- terial invasion of gingiva in advanced periodontitis in humans. I Periodontol 1981: 22: 217-222.

256. Salvi GE, Beck ID, Offenbacher S. PGE., IL-1 beta, and TNF-alpha responses in diabetics as modifiers of periodontal disease expression. Ann Periodontol 1998: 3: 40- 50.

257. Salvi GE, Collins JG, Yalda B, Arnold RR, Lang NP, Offen- bacher S. Monocytic TNFa secretion patterns in IDDM patients with periodontal diseases. J Clin Periodontol 1997: 24: 8-16.

258. Salvi GE, Yalda B, Collins JG, Jones BH, Smith FW, Arnold RR, Offenbacher S. Inflammatory mediator response as a potential marker for periodontal diseases in insulin-de-

pendent diabetes mellitus patients. J Periodontol 1997: 68: 127-135.

259. Sanchez L, Calvo M, Brock J. Biological role of lactoferrin. Arch Dis Child 1992: 67: 657-661.

260. Sandholm L, Saxen L. Local immunoglobulin synthesis in juvenile and adult periodontitis. J Clin Periodontol 1984: 11: 459-466.

261. Sandholm L, Tolo K, Olsen 1. Salivary IgG. A parameter of periodontal disease activity? J Clin Periodontol 1987: 14: 289-294.

262. Santos R, Shanfeld J, Casamassimo P Serum antibody response to Actinobacillus actinomycetemcomitans in Down's syndrome. Spec Care Dent 1996: 16: 80-83.

263. Saxby MS. Juvenile periodontitis: an epidemiological study in the west Midlands of the United Kingdom. J Clin Periodontol 1987: 14: 594-598.

264. Schenkein H, Van Dyke T. Early onset periodontitis. Sys- temic aspects of etiology and pathogenesis. Periodontol 2000 1994: 6: 7-25.

265. Schenkein HA, Genco RJ. Gingival fluid and serum in periodontal diseases. if. Evidence for activation of complement components C3, C3 proactivator and C4 in gingival fluid. J Periodontol 1977: 48: 778-784.

266. Schiött C, Löe H. The origin and variation in number of leukocytes in the human saliva. J Periodontal Res 1970: 5: 36-41.

267. Schroeder H, Graf-de-beer M, Attström R. Initial gingivitis in dogs. J Periodontal Res 1975: 110: 128-142.

268. Schroeder H, Munzel-Pedrazzoli S, Page R. Correlated morphometric and biochemical analysis of gingival

33

Kinane et al.

tissues in early chronic gingivitis in man. Arch Oral Biol 1973: 18: 899-923.

269. Scully C, Challacombe S. The migration of 111 indium- labeled polymorphonuclear leukocytes into the oral cavity in the rhesus monkey. I Periodontal Res 1979: 14: 475-481.

270. Sela MN, Kornman KS, Ebersole IL. Holt SC. Characteriza-

tion of treponemes isolated from human and non-human primate periodontal pockets. Oral Microbiol Immunol 1987: 2: 144-149.

271. Seymour G, Powell R, Davies W. Conversion of a stable T-

cell lesion to a progressive B cell lesion in the pathogenesis of chronic inflammatory periodontal disease: an hypothesis. J Clin Periodontol 1979: 6: 267-275.

272. Seymour GJ, Crouch MS, Powell RN, Brooks D, Beckman I, Zola H, Bradley J, Burns GE The identification of lymph-

oid cell subpopulations in sections of human lymphoid

tissue in gingivitis in children using monoclonal antibodies. J Periodontal Res 1982: 17: 247-256.

273. Seymour GJ, Powell RN, Cole KL, Aitken JF, Brooks D, Beckman I, Zola H, Bradley J, Burns GE Experimental gingivitis in humans. A histochemical and immunological

characterisation of the lymphoid cell subpopulations. I Periodontal Res 1983: 18: 375-385.

274. Shapira L, Soskolne WA, Sela MN, Offenbacher S, Barak V. The secretion of PGE2, IL-1ß, IL-6, and TNFa by adherent

mononuclear cells from early onset periodontitis patients. I Periodontol 1994: 65: 139-146.

275. Sharon J. Basic immunology. Baltimore: Williams & Wilk- ins, 1998.

276. Sharry II, Krasse B. Observations on the origin of salivary leukocytes. Acta Odontol Scand 1960: 18: 347-358.

277. Sims TJ, Moncla BI, Darveau RP, Page RC. Antigens of Actinobacillus actinomycetemcomitans recognized by patients with juvenile periodontitis and periodontally normal subjects. Infect Immun 1991: 59: 913-924.

278. Skapski H, Lehner T. A crevicular washing method for in-

vestigating immune components of crevicular fluid in

man. J Periodontal Res 1976: 11: 19-24.

279. Sklar LA, Onman GM. Kinetics and amplification in neutrophil activation and adaptation. Semin Cell Biol 1990: 1: 115-123.

280. Slots J, Genco RI. Black pigmented bacteroides species,

capnocytophaga species and actinobacillus actinomycetemcomitans in human periodontal disease: virulence factors in colonisation, survival and tissue destruction. I

Dent Res 1984: 63: 412-421.

281. Smibert RM, Burmeister JA. 7i'eponema pectinovarum sp. nov. isolated from humans with periodontitis. Intl I Syst Bacte rio l 1983: 33: 852-856.

282. Smibert RM, Johnson JL, Ranney RR. 7teponema socranskii sp. nov.; Treponema socranskii subsp. socranskii subsp. nov; Treponema socranskii susp. buccale subsp. nov.; and 7)-eponema socranskii subsp. paredis subsp. nov. isolated from humans with experimental gingivitis and periodontitis. Intl J Syst Bacteriol 1984: 34: 456-462.

283. Smith DJ, Gadalla LM, Ebersole JL, Taubman MA. Gingival crevicular antibody to oral microorganisms. III. Associ- ation of gingival homogenate and gingival crevicular fluid antibody levels. J Periodontal Res 1985: 20: 357-367.

284. Snyderman R, Pike MC. Chemoattractant receptors on phagocytic cells. Ann Rev Immunol 1984: 2: 257-281.

285. Sorsa T, Suomalainen K, Uitto VJ. The role of gingival crevicular fluid and salivary interstitial collagenases in hu-

man periodontal diseases. Arch Oral Biol 1990: 35: 193S- 196S.

286. Sorsa T, Uitto VJ, Suomalainen K, Vauhkonen M, Lindy S. Comparison of interstitial collagenases from human gingiva, sulcular fluid and polymorphonuclear leukocytes. J Periodontal Res 1980: 23: 386-393.

287. Sperber GH. Oral contraceptive hypertrophic gingivitis. I Dent Assoc S Afr 1969: 24: 37-40.

288. Stashenko P, Dewhirst FE, Rooney ML, Desjardins LA, Heeley JD. Interleukin-lp is a potent inhibitor of bone for-

mation in vitro. J Bone Miner Res 1987: 2: 559-565. 289. Stashenko P Fujiyoshi P, Obernesser MS, Prostak L, Haf-

fajee AD, Socransky SS. Levels of interleukin-1ß in tissue from sites of active periodontal disease. J Clin Periodontol 1991: 18: 548-554.

290. Stashenko P, Jandinski JJ, Fujiyoshi P, Rynar I, Socransky SS. Tissue levels of bone resorptive cytokines in periodontal disease. J Periodontol 1991: 62: 504-509.

291. Staunton DE, Dustin ML, Erickson HP, Spinger TA. The arrangement of the immunoglobulin like domains of ICAM-1 and the binding sites for LFA-1 and rhinovirus. Cell 1990: 61: 243-254.

292. Stein SH, Hart TE, Hoffman WH, Hendrix CL, Gustke CJ, Watson SC. Interleukin-10 promotes anti-collagen antibody production in type I diabetic peripheral B lymphocytes. I Periodontal Res 1997: 32: 189-195.

293. Stein SH, Hendrix CL. Interleukin-10 promotes anti-collagen antibody production in gingival mononuclear cells. I Dent Res 1996: 75: 158 (abstr).

294. Steubing PM, Mackler BF, Schur PH, Levy BM. Humoral

studies of periodontal disease 1. Characterization of im-

munoglobulins quantitated from cultures of gingival tissue. Clin Immunol Immunopathol 1982: 22: 32-43.

295. Stoufi ED, Taubman MA, Ebersole JL, Smith DI, Stashenko PP. Phenotypic analyses of mononuclear cells recovered from healthy and diseased human periodontal tissues. J Clin Immunol 1987: 7: 235-245.

296. Stoufi ED, Taubman MA, Ebersole JL, Smith DI. A method for studying immunoglobulin synthesis by gingival cells. Oral Microbiol Immunol 1988: 3: 109-112.

297. Sugawara M, Yamashita K, Yoshie H, tiara K. Detection of, and anti-collagen antibody produced by, CD5-positive B cells in inflamed gingival tissue. J Periodontal Res 1992: 27: 489-498.

298. Suomalainen K, Sorsa T, Saxen L, Vauhkonen M, Uitto VI. Collagenase activity in gingival crevicular fluid of patients with juvenile periodontitis. Oral Microbiol Immunol 1991: 6: 24-29.

299. Sutcliffe P. A longitudinal study of gingivitis and puberty. I Periodontal Res 1972: 7: 52-58.

300. Syrjanen S, Markkanen EI, Syrjanen K. Inflammatory cells

and their subsets in lesions in juvenile periodontitis. A family study. Acta Odontol Scand 1984: 42: 285-292.

301. Taichmann NS, Tsai CC, Baehini PC, Stoller N, McArthur WP. Interaction of inflammatory cells and oral micro-organisms. IV. In vitro release of lysosomal constituents from polymorphonuclear leukocytes exposed to supragingival and subgingival bacterial plaque. Infect Immun 1977: 16: 1013-1023.

302. Takahashi K, Poole 1, Kinane DF. Detection of interleukin-

IjI mRNA-expressing cells in human gingival revicular fluid by in situ hybridization. Arch Oral Biol 1995: 40: 941- 947.

34


303.

304.

305.

306.

307.

308.

309.

310.

311.

312.

313.

314.

315.

Takahashi K, Takigawa M, Takashiba S, Nagai A, Miyamo-

to M, Kurihara 11, Murayama Y. Role of cytokine in the induction of adhesion molecules on cultured human gingival fibroblasts. J Periodontol 1994: 65: 230-235. Takashiba S, Takigawa M, Takahashi K, Moyokai F, Nishi-

mura F, Chihara T, Kurihara H, Nomura Y, Murayama Y. Interleukin-8 is a major neutrophil chemotactic factor de-

rived from cultured human gingival fibroblasts stimulated

with interleukin-lbeta or tumor necrosis factor alpha. In- fect Immun 1992: 60: 5253-5258. Takiguchi H, Yamaguchi M, Okamura 11, Abiko Y. Contri- bution of IL-Ibeta to the enhancement of Camplylobacter

rectos lipopolysaccharide-stimulated PGE2 production in

old gingival fibroblasts in vitro. Mech Ageing Dev 1997: 98: 75-90.

Tangada SD, Califano JV, Nakashima K, Quinn SM, Zhang JB, Gunsolley JC, Schenkein HA, Tew JG. The effect of smoking on serum lgG2 reactive with Actinobacillus actinomycetemcomitans in early-onset periodontitis patients. J Periodontol 1997: 68: 842-850. Tanner ACR, Haffer C, Bratthall GT, Visconti RA, Socran-

sky SS. A study of the bacteria associated with advancing

periodontitis in man. J Clin Periodontol 1979: 6: 278-307.

Taubman MA, Stoufi ED, Ebersole JL, Smith DJ. Pheno-

typic studies of cells from periodontal disease tissues. 1


Tew JG, Marshall DR, Moore WEC, Best AM, Palcanis KG,

Ranney RR. Serum antibody reactive with predominant organisms in the subgingival flora of young adults with generalized severe periodontitis. Infect Immun 1985: 48: 303-311. Tew JG, Zhang 113, Quinn S, Tangada S, Nakashima K, Gunsolley JC, Schenkein HA, Califano N. Antibody of the IgG2 subclass, Actinobacillus actinomycetemcomitans,

and early-onset periodontitis. J Periodontol 1996: 67: 317- 322. Thomas EL, Lehrer RI, Rest RE Human neutrophil anti-

microbial activity. Rev Infect Dis 1998: 10(suppl 2): S450-

S456. Tonetti MS, Mombelli A. Early onset periodontitis. In:

Lindhe I, Karring T, Lang NP, ed. Clinical periodontology and implant dentistry. 3rd edn. Copenhagen: Munks-

gaard, 1997: 226-257.

Tonetti MS. Etiology and pathogenesis. In: Lang NP, Karring T, ed. Proceedings of the Ist European Workshop on Periodontology. London: Quintessence Publishing Company. 1993: 54-89. Tsai C-C, Chen C-C, llou G-L. Serum antibody against Actinobacillus actinornycetemcornitans in Chinese patients with periodontitis. Kaohsiung J Med Sci 1987: 3: 714-722.

Tsai CC, Taichman NS, Genco RI, Evian Cl, Baehni PC, McArthur WP. Serum neutralising activity against Actino- bacillus actinornycetemcomitans leukotoxin in juvenile periodontitis. I Clin Periodontol 1981: 8: 338-348. tJitto V-J, Turto 1-1, Saxen L. Extraction of collagenase from human gingiva. J Periodontal Res 1978: 13: 207-214. Van der Weijden GA, Timmerman MF, Danser MM, Nijbo-

er A, Saxton CA, Van der Weiden U. Effect of pre-experimental maintenance care duration on the development

of gingivitis in a partial mouth experimental gingivitis model. I Periodontal Res 1994: 29: 168-173.

Van der Zee E, Everts V, Beertsen W. Cytokines modulate

316.

317

318.

routes of collagen breakdown. Review with special emphasis on mechanisms of collagen degradation in the periodontium and the burst hypothesis of periodontal disease progression. J Clin Periodontol 1997: 24: 297-305.

319. Van Dyke TE, Horoszewicz HU, Cianciola LI, Genco RJ. Neutrophil chemotaxis dysfunction in human periodontitis. Infect Immun 1980: 27: 124-132.

320. Van Dyke TE, Horoszewicz HU, Genco RI. The polymorphonuclear leukocyte (PMNL) locomotor defect in juvenile periodontitis. I Periodontol 1982: 53: 682-687.

321. Van Dyke TE, Offenbacher S, Kalmar I. Arnold RR. Neu- trophil defects and host-parasite interactions in the pathogenesis of localised juvenile periodontitis. Adv Dent Res 1988: 2: 354-358.

322. Van Dyke TE, Schenkein HA. Research objectives for the study of early-onset periodontitis. A summary of the working groups for the early-onset periodontitis workshop. J Periodontol 1996: 67(suppl): 279-281.

323. Van Dyke TE, Schweinebraten M, Cianciola LJ, Offen- bacher S, Genco RI. Neutrophil chemotaxis in families with localized juvenile periodontitis. I Periodontal Res 1985: 20: 503-514.

324. Van Dyke TE, Wilson-Burrows C, Offenbacher S, Henson P. Association of an abnormality of neutrophil chernotaxis in human periodontal disease with a cell surface protein. Infect Immun 1987: 55: 2262-2267.

325. Van Dyke TE, Zinney W, Winkel K, Taufiq A, Offenbacher S, Arnold RR. Neutrophil function in localised juvenile periodontitis: phagocytosis, superoxide production and specific granule release. J Periodontol 1986: 57: 703-708.

326. Van Winkelhoff Al, Van der Velden U, De Graaff J. Mi- crobial succession in recolonizing deep periodontal pockets after single course of supra- and subgingival debridement. I Clin Periodontol 1988: 15: 116-122.

327. Vandesteen GE, Page RC, Altman LC, Ebersole JL, Willi- ams BL. Clinical, microbiological, immunological studies of a family with a high prevalence of early-onset periodontitis. I Periodontol 1984: 55: 159-169.

328. Verhoef J, Visser MR. Neutrophil phagocytosis and killing:

normal function and microbial evasion. In: Abramson IS, Wheeler JG, ed. The neutrophil. New York: Oxford Univer-

sity Press, 1993: 109-137.

329. Villela B, Cogen R, Bartolucci A, Birkedal-Hansen H. Col- lagenolytic activity in crevicular fluid from patients with chronic adult periodontitis, localised juvenile periodontitis and gingivitis, and fror, i healthy control subjects. J Periodontal Res 1987: 22: 381-389.

330. Villela B, Cogen R, Bartolucci A, Birkedal-Hansen H. Cre- vicular fluid collagenase activity in healthy, gingivitis, chronic adult periodontitis and localised juvenile periodontitis patients. J Periodontal Res 1987: 22: 209-211.

331. Vincent JW, Suzuki JB, Falkler WA in, Cornett WC. Reaction of human sera from juvenile periodontitis, rapidly progressive periodontitis, and adult periodontitis patients with selected periodontopathogens. I Periodontol 1985: 56: 464-469.

332. Vittek 1, Hernandez MR, Wenk EJ, Rappaport SC, South- ern AL Specific estrogen receptors in human gingiva. J Clin Endocrinol Metabol 1982: 54: 608-612.

333. Vittek 1, Munnangi PR, Gordon G, Rappaport SC, South- ern AL Progesterone "receptors" in human gingiva. IRCS Med Sci 1982: 10: 381-390.

334. Vittek I, Rappaport SC, Gordon GC, Munnangi PR, South-

35

Kinane et al.

ern AL. Concentration of circulating hormones and met-

abolism of androgens by human gingiva. I Periodontol

1979: 50: 254-264.

335. Waldrop T, Anderson D, Hallmon W, Schmalstieg F, Jacobs

R. Periodontal manifestation of the heritable Mac-1, LFA-

1, deficiency syndrome. 1 Periodontol 1986: 58: 400-416.

336. Waldrop TC, Mackler BF, Schur P. 1gG and IgG subclasses in human periodontitis (juvenile periodontitis). Scand J

Dent Res 1981: 86: 421-429.

337. Wallace P, Packman C, Lichtman M. Maturation associated changes in the peripheral cytoplasm of human neutrophils: a review. Exp Haematol 1987: 15: 34-45.

338. Watanabe it, Marsh PD, Ivanyi L. Antigens of Actino-

bacillus actinomycetemcomitans identified by immuno-

blotting with sera from patients with localized human

juvenile periodontitis and generalized severe peri-

odontitis. Arch Oral Biol 1989: 34: 649-656.

339. Watanabe H, Umeda M, Seid T, Ishikawa I. Clinical and laboratory studies of severe periodontal disease in an

adolescent associated with hypophosphatasia. A case re-

port. 1 Periodontol 1993: 64: 174-180. 340. Weiss 1, Victor M, Stendahl 0, Elsbach P. Killing of gram

negative bacteria by polymorphonuclear leukocytes. Role

of an O2-independent bactericidal system. I Clin Invest

1982: 69: 959-970. 341. Whicher IT, Evans SW. Cytokines in disease. Clin Chem

1990: 36: 1269-1281.

342. Wilson ME, Hamilton RG. Immunoglobulin G subclass re-

sponse of juvenile periodontitis subjects to principal outer membrane proteins of Actinobacillus actinomycetemcomitans. Infect Immun 1995: 63: 1062-1069.

343. Wilson ME, Kalmar JR. EcRlla (CD32) -a potential marker

defining susceptibility to localized juvenile periodontitis.

I Periodontol 1996: 67: 323-331. 344. Wilson ME, Schifferle RE. Evidence that the serotype b

antigenic determinant of Actinobacillus actinomycetem-

comitans Y4 resides in the polysaccharide moiety of lipo-

polysaccharide. Infect Immun 1991: 59: 1544-1551.

345. Wilson ME. IgG antibody response of localized juvenile

periodontitis patients to the 29 kilodalton outer membrane protein of Actinobacillus actino mycetemcom i tans. I

Periodontol 1991: 62: 211-218. 346. Wilton 1MA, lvanyi L, Lehner T. Cell-mediated immunity

and humoral antibodies in acute ulcerative gingivitis. I


3,17. Wojcicki C1, Harper DS, Robinson Pl. Differences in peri-

odontal disease-associated microorganisms of subgingival plaque in prepubertal, pubertal and postpubertal children. I Periodontol 1987: 58: 219-223.

348. Woolley 1), Davies R. Immunolocalization of collagenase in periodontal disease. I Periodontal Res 1981: 16: 292-

297.

349. Wright DG, Gallin IL Secretory responses of human neu-

trophils. Exocytosis of specific (secondary) granules by

human neutrophils during adherence in vitro and during

exudation in vivo. I Immunol 1979: 123: 285-294.

350. Wright SD, Tobias PS, Ulevitch RJ, Ramos RA. Lipopoly-

saccharide (LPS) binding protein opsonises LPS bearing

particles for recognition of a novel receptor on macro-

phages. J Exp Med 1989: 170: 1231-1241. 351. Wynne SE, Walsh Li, Seymour GJ, Powell RN. In situ dem-

onstration of natural killer (NK) cells in human gingival

tissue. I Periodontol 1986: 57: 699-702.

352. Yamamoto M, Nishihra T, Koseki T, He T, Yamato K, Zhang YJ, Nakashima K, Oda S, Ishikawa 1. Prevalence of Actino- bacillus actinomycetemcomitans serotypes in Japanese

patients with periodontitis. Oral Microbiol Immunol 1997: 32: 676-681.

353. Yamamura M, Uyemura K, Deans RJ, Weinberg K, Rea TH,

Bloom BR, Modlin RL. Defining protective responses to

pathogens: cytokine profiles in leprosy lesions. Science

1991: 254: 277-279.

354. Yamazaki K, Nakajima T, Aoyagi T, Hara K. Immunohistol-

ogical analysis of memory T lymphocytes and activated B lymphocytes in tissues with periodontal disease. J Peri-

odontal Res 1993: 28: 324-334. 355. Yamazaki K, Nakajima T, Hara K. Immunohistological

analysis of T cell functional subsets in chronic inflamma-

tory periodontal disease. Clin Exp Immunol 1995: 99:

384-391. 356. Yamazaki K, Nakajima T, Kubota Y, Gemmell E, Seymour

GJ, Hara K. Cytokine messenger RNA expression in

chronic inflammatory periodontal disease. Oral Microbiol

Immunol 1997: 12: 281-287.

357. Yavuzyilmaz E, Yamalik N, Bulut S, Uzen S, Ersoy F, SaatFi

U. The gingival crevicular fluid interleukin-1ß and tumour

necrosis factor-a levels in patients with rapidly progressive periodontitis. Austr Dent J 1995: 40: 46-49.

358. Zafiropoulos GG, Flores-de-Jacoby L, Schoop B, Havem-

ann K, Heymanns J. Flow cytometric analysis of lympho-

cyte subsets in patients with advanced periodontitis. I

Clin Periodontol 1990: 17: 636-641.

359. Zambon JI, Christersson LA, Slots I. Actinobacillus actinomycetemcomitans in human periodontal disease: prevalence in patient groups and distribution of biotypes and serotypes within families. J Periodontol 1983: 54: 707-711.

360. Zambon JJ, Haraszthy VI, Hariharan G, Lally ET, Demuth DR. The microbiology of early onset periodontitis: association of highly toxic Actinobacillus actinomycetemcomitans strains with localized juvenile periodontitis. J Peri- odon to l 1996: 67: 282-290.

361. Zambon II. Periodontal diseases: microbial factors. Ann Periodontol 1996: 1: 879-925.

362. Ziskin DE, Blackberg SN, Slanetz CA. Effects of subcuta-

neous injections of estrogenic and gonadotrophic hor-

mones on gums and oral mucous membranes of normal and castrated rhesus monkeys. J Dent Res 1936: 15: 407- 428.

, ýi,;:. `

36

Documents

Download (29Mb) - Glasgow Theses Service - University of Glasgow